We have a proper treat this week. Core Memory special correspondent Eryney Marrogi comes on the pod.
Marrogi is a scientist and soon-to-be doctor who has followed the rise of the gene editing tools at humanity’s disposal. He wrote a piece for us a couple of months ago on “Baby KJ,” the child who received a customized, life-saving gene therapy in record time. Now we get into what Baby KJ means for the future of gene therapies.
The gene therapy field is advancing quickly but remains costly and technically complex. Marrogi breaks down how hopeful people should be – or not – that new techniques can be applied to more than super rare, single gene problems.
We also get into the ups and downs of the DIY Bio movement, Marrogi’s work creating gene-edited mosquitoes and the vaping crisis he sees with America’s youth, who are taking in unfathomable levels of nicotine.
Enjoy!
The Core Memory podcast is sponsored by Brex, the intelligent finance platform. Like thousands of ambitious, innovative companies, we run on Brex so we can spend smarter and move faster. And you can too. Learn more at www.brex.com/corememory
The podcast is also made possible by E1 Ventures, which backs the most ambitious founders and start-ups.
No summary for this episode yet.
Speaker 3
San Francisco has like too many billboards of like... On the freeway. Yeah, like agentic workflows for like things I didn't know that you could have a workflow for.
Speaker 1
Yeah. Yeah. Yeah, well, I mean, this is the great San Francisco tradition is the billboards on the 101 where you... understand what is the next new thing. No, yeah, yeah.
Speaker 3
I'm hoping to pay pilgrimage to the Clulee billboard, wherever that is. I would like to see that with my eyes, but I don't know. It might not happen this time. I felt like that was working for him for a week, and now it's... It turns out, yeah, the... The amount of goodwill you can get on Twitter is pretty limited.
Speaker 1
They need a new shtick. Yeah, the billboards are funny, though, because you can... I always use this as my hype cycle gauge. Oh, yeah. Because you go... It's like after when the AI thing started to take off, it's like, you know, there's two or three, and then...
Speaker 3
Now, I didn't see a single injury law firm. No. Not a single one.
Speaker 1
It's like every billboard is an AI billboard. And then, like you said, if you're not from here. It's jarring. Right, because it's not just like, oh, it's OpenAI and Claude. Yeah, yeah. Super idiosyncratic AI.
Speaker 3
This is like shoe.ai.
Speaker 1
We do AI shoes, whatever that means. I don't know if they still have it, but right as you're getting into San Francisco, it's like renting GPUs and stuff. Like, where do we live, man? SF Compute.
Speaker 3
Yeah, yeah, yeah. It's good. I like it. I mean, Vermont doesn't have any billboards. There's zero billboards in Vermont. It's like a law. So to go from like that to this, I'm like, wow, this is... yeah it's it's very jarring i think there's one that's like pro israel and then right behind it is like pro no i think that they put like a pro palestinian one up and then some israeli group put the right behind well at least they're balanced it's like a balance like a yin and yang type thing yeah it's good to have balance
Speaker 1
with these things um okay we'll jump in you're gonna I want to get your, I'm going to take a crack. You're going to have to help me on the last thing. Okay, yeah, yeah. Do you want me to do it for you? Let me, you grade me. Okay. Erne Moroghi.
Speaker 3
Aaron A. Morogi. Aaron A. Aaron A. Aaron A. It's very simple. It's like Aaron A. Very simple.
Speaker 1
Aaron A. And I listened to this a bunch of times. No, it's okay. It's okay. Aaron A. Marogi.
Speaker 3
Yeah.
Speaker 1
But the O is almost a U. Marogi is good.
Speaker 3
Yeah. Marogi is good. Yeah. Where's, where's, uh, yeah. So yeah, my mom is from, is Ukrainian. Um, and my dad is from Iraq. Okay. And so the last name is Iraqi. I guess they're like the first name is Ukraine. If I want to simplify it, it's like Ukrainian. We'll do that. Um, but yeah, so yeah. Do you get earned a lot? yeah I get it all yeah so like a very common thing is like I'll introduce myself to like a patient in the hospital I'm Aaron A I'm like working on your medical team and they're like okay Aaron it's like it's fine like I it's I just let it go but um I get Ernie I get Henry like I get it all whatever you think I could get it I've gotten it well online I am a woman yeah I'm going to be honest. Until like a year ago, I was like, yeah, this Ashley woman, she's just tapped into everything. And I had like, I had read an article that you wrote and I feel like finally I was like, I'm going to like, what is this woman's like, How did she get into this? And then I obviously corrected.
Speaker 1
It sort of broke my brain a little bit. This lets me know my video show is not doing well enough. No, no, no. No, no, because this happens all the time. People are like, I've followed you for 12 years. I'm like, oh, have you never seen like one video?
Speaker 3
Well, I don't know. Like I read, like I don't watch as much stuff. Like I read a lot of stuff. Yeah. And I'm also like a little, like I'm online. Like I'm online like compared to like, the median American but compared to like the SF cluster like I'm a left tail for sure in terms of so like I'll read things like I was subscribed to Core Memory read like Bloomberg stuff but like I didn't Yeah, I never watched... I'll be honest.
Speaker 1
I was not watching the documentary. Not a requirement. It's pretty funny. When I used to write for The Register, which is a British tech publication that's quite geeky, and I covered semiconductors and data-centered gear, I think I was many dudes' fantasy because I'm writing about just the nerdiest stuff that I would get. I would get these late-night emails and be like, hey you want to get a drink yeah are you coming to like north carolina anytime soon it's like i would love to talk about risk architecture with you and uh And then, and then I remember I got one of those and then like 10 minutes later the dude was like, oh God, I just Googled you. He's like, I did an image search. I'm, I'm so sorry. And then, and then it was, it was like a little insight into, um, life as a female writer because I would also, the tone of a lot of it would, it would also be like, uh, hey, sweetheart, you're a moron. You know, like stuff that I did not get from people who knew. Yeah. Yeah. And I was like, oh, God, okay. Yeah, yeah. It's like you guys are just, you just talk down to me, you know, like in a pretty aggressive way right off the bat. Yeah.
Speaker 3
I will say you don't have a picture on your Twitter. So it's like just a core memory. Which, respect that. I respect that. But it certainly doesn't help you. I like to be a man of a stranger.
Speaker 1
All right, what do we do at Core Memory? We cover innovative, fast-moving, forward-thinking companies, which is why Core Memory is sponsored by Brex, because Brex is the intelligent finance platform for many of these companies, 30,000 companies from startups to the world's largest corporations. rely on brex's technology for their finances they've got smart corporate cards high yield business banking and expense automation tools that are fantastic i hate doing my expenses and brex's ai's and software run right through those expenses figure out where we're spending money and take care of so much stuff for you so you don't have to waste your time on it yourself Go to brex.
com slash core memory to learn more and just, you know, get with the program. Let's get going. Let's get out of this archaic finance software and move toward the future. Core memory and Brex. Well, let me, let me tell people who you are. Um, one reason, okay, well, you're a medical, you're, you're pursuing, you're going to be a doctor. Yeah. But you're a, you're a scientist. Um, you have written well okay you're multi-faceted i mean you used to do research at places like george church's lab kind of at the cutting edge of what people sometimes describe as diy bio really learning how to turn biology into computation this this where these two things intersect and and then you do some scientific writing Quite notably for us, you just wrote this wonderful piece for Core Memory on Baby KJ, which is I think where I want to start, but then I want to get into your career and your avenues of research and where you think the world is heading.
But yeah, I mean, is that kind of fair for a bit of an intro to you?
Speaker 3
Yeah, I mean, I think that's totally fair. Yeah, I had like a scientific career, and then I decided, I went to go do medicine, so I'm medical school now. And I do, like I've been writing along the way, but it really picked up in like 2023. And yeah, so I mean, like with the story, first, I'm very glad to have been able to write about this. I guess if we could take a little bit of a walk, I think... You know, where my mind has kind of been with the gene therapy stuff, it really started when I was in Georgia's lab. I, you know, the church lab is this really wonderful, almost like fantastical place where people work on whatever science fiction stuff you can, like in your mind, people work on crazier stuff than that.
Speaker 1
Yeah. So just super quick for people who have no idea who George, I mean, world famous scientist, like you said, always seems to be the guy who's cloning something, bringing animals back. He's been doing the mammoth project to try to bring back the woolly mammoth, always, always at the kind of edge. And then, and then I'm going to shut up in one second, but totally during my career. He's fascinating. And then, and then so many students that go through his lab and he's all, he's co-founded like 8,000 startups from all these students through, through his lab. So yeah, just, just kind of this very, um, eccentric, well-known figure.
Speaker 3
Yeah, absolutely. Yeah. And I think. Yeah, so, you know, the lab, he kind of, I think these days, most people think, you know, if you Google George, like, you might find a lot of stuff about, like, the mammoth, the woolly mammoth stuff. I think it's really picked up a lot of steam. But the reality is, like, George has been doing a ton of just... Bleeding Edge is not even, like he's setting the frontier of biology. And he's been doing that for a long, long, long time. And his lab has really been a place for people to go to, to work on like crazy, kind of like really ambitious, scientifically ambitious, but really science fiction type stuff.
And, you know, we'll talk about KJ in a second. But I think like, for instance, if I think about my first... week in Georgia's lab, I went to a lab meeting. So the lab is really large. At the time when I was there, it was like 150 people split between Harvard and MIT. The Wiese Institute is kind of like a multi-university sort of systems biology or like, you know, biological engineering institution. Uh, and the lab is really big. So they split the, the, the lab meetings. You have to have two every week. They're like an hour long, Tuesdays and Thursdays. And the first lab meeting I went to was given by a guy, his name is Seth Shipman.
He's currently a professor at UCSF. This is like 2016. And he gave a presentation where they put, they made like a bacteria, like hold, uh, like a GIF of a horse running, like, uh, We can do that? You can just make a bacteria be a flash drive.
Speaker 1
That's crazy. What does that even mean? Just the ones and zeros from this?
Speaker 3
Yeah, I mean, there's a little more complication to it. And that works. I watched this presentation. This is 2016. I won't even try to describe it because... You know, I'll butcher it. And I don't want to disrespect what Seth has done there. But, you know, it ended up being, like, a science paper. It was, like, a big deal when they did it. Because it's just, like, it's an insane concept. And now people have been trying to turn these concepts into, like, companies, which is, can we make, you know, DNA stored data, right? Like, we have data shortages, whatever. But in any case, like, where I'm going with this is that, like, Georgia's lab was this place where people were working on super ambitious things.
And, you know, these days, and when I was there, the big thing was CRISPR. Like CRISPR-Cas9 was this monumental change in what we could start to imagine as possible with biology. Obviously, the story of CRISPR-Cas9 is not one that originated in Georgia's lab, you know. Sort of leave the debate there. We don't have eight hours. Yeah, yeah, we won't do it. But, you know, Jennifer Doudna's, like, you know, at Berkeley, they were the ones who were, like, showing there's a system, an existing bacteria, and, oh, actually, like, maybe we can do something interesting with this. And then George and Fung at the Bird Institute, so Harvard and MIT, showed, actually, we can apply this system to mammalian cells.
Turns out we actually, we can edit human cells. Yeah. And that's like something people have talked about. Like gene therapy has been this idea for, I mean, probably since the 80s, 90s, when we started to understand genetics, we started to understand that there are certain diseases that are driven by aberrant mutations in our DNA, like the code that makes us run. can sometimes come messed up or it can become messed up over time. And with CRISPR-Cas9, all of a sudden, the idea of what we could do with that was very different. Before that, you had a disease, maybe you had cancer, maybe you had something else, and you try to find a small molecule.
And a small molecule, you don't really know. Maybe sometimes you don't really know exactly what it does, but you know it improves the condition or it doesn't improve the condition. But with CRISPR-Cas9, all of a sudden, the idea is somebody has a disease, And we can figure out exactly what is causing that disease. So maybe it's a specific mutation in the DNA, and this is making like a broken protein. And that broken protein is making somebody very sick. And, oh, we can actually just reverse that. We can just fix it. And now all of a sudden, the pitch was like a one-time thing.
Speaker 1
Right. Instead of perpetually treating it through a drug, you just get rid of it.
Speaker 3
Yeah, exactly. You know, that's been the idea. I think there's some smoothing out that needs to happen, but that was really the idea. And so, you know, I've been working around that idea for a long time. I really bought into this concept of like, can we give people like agency over who they are, like their DNA, right? It drives so much of... what it means to be, you know, Ashley or Erin. There are things that are in our DNA that are written there, and it kind of dictates a lot about us. And unfortunately for some people, it means that they have to deal with pretty serious conditions from birth and onwards.
And so, CRISPR-Cas9 has had this huge, you know, there was the original thing, but then all of a sudden, people started to work on like better ways of using the system. And we'll, we'll talk about, you know, David Lu has been a big part of the KJ story. Um, but you know, if I had to sort of summarize the whole thing, it's like CRISPR-Cas9 all of a sudden allowed us to really imagine, uh, just going way upstream, like what causes the disease? Let's just go all the way up to what started it and just fix it. And now you're done. And that was kind of like what the story to some extent was with KJ.
Speaker 1
Yeah. I mean, this is a perfect lead in and I want to, I want to, I have more questions on some of the stuff you said, but let's, let's unpack KJ for people who might not have known, you know, a few months ago, I think it was the front page of the New York times or it was in the New York times that, Hey, there was this, this baby, um, in, in Philadelphia, in Pennsylvania and, and got known as baby KJ and, One of these children that you read about quite often that has this rare genomic disease, and it's, you know, like you said, sometimes it's one gene, sometimes it's a few, but something goes haywire, and these conditions tend to be quite awful.
I mean, it's something that... is likely to drastically shorten the lifespan of these children and and or to lead to like a very difficult life that where you required constant treatment um and and often we don't have treatment so people have been spending Millions of dollars. It's not a very good market for therapies. So the pharma companies don't spend a lot of time on this because you're only going to be treated, even though there are in total millions of children with these conditions, each one is too small to like spend the R and D on. And so, yeah, we've had this hope that, oh, well, maybe you could do gene therapy.
It's a touchy area because CRISPR, even though it's been around for a while, remains relatively new in the big scheme of things. And it's like, oh, are you going to start meddling with this child's DNA? And is that ethical? Is it safe? All these questions. And so Baby KJ was this, I think, It's fair to say, right? Like the first example of this child was born, he has this terrible condition and we're going to take this risk and do this treatment. And then it seems like it's working. And so, yeah. I mean, tell us the story. What did he have? What did we know going in? And yeah, how is this different to what had come before?
Speaker 3
Yeah, so, you know, there's so many avenues we can go here, but we'll sort of start with the basics. So, like, last November, a baby was born in Philadelphia, K.J. Muldoon. You know, sort of, you know, family, very, you know, loving parents, baby was born, nobody was really thinking there was anything going awry here. But a couple days in, all of a sudden, things are not going so hot for K.J. You know, he's having... some metabolic dysfunction. And there are kind of a lot of things that we think about in medicine when a baby's born and okay, maybe they're jaundiced, maybe they're having a fever, maybe they're not able to fight off infection.
There are things that we think about, but kind of where they went at CHOP, so Children's Hospital of Philadelphia, where they went, it was like, actually maybe this is like a genetic condition. And in KJ's case, what he was born with is a CPS1 deficiency. So what does that actually mean? Your body has a way of dealing with protein. So you eat protein, and then some of it gets digested. And there's a cycle involved in that. We call it the urea cycle. And that's our body's way of making certain amino acids, dealing with other amino acids. But in KJ's case, one of the enzymes in that loop was broken.
CPS1. And in his case, you know, he's born with this mutation. He has a couple mutations, really, but like very specific amino acids that are broken. And it's a very rare disease, one in, I think, one and a half million. So it's rather rare. And as you were sort of alluding to, there are other enzymes that can break in that cycle. You know, there are several enzymes involved in it. And any one of them can break. And they all kind of have the same end phenotypes, the same sort of end manifestation of the disease, which is like, you can't deal with protein. And so KJ is born, they kind of do some lab tests, they determine his buildup of like these metabolites, which is really bad, and it's causing him to have a ton of ammonia in his blood.
And that's uh can be deadly and if it's not deadly a lot of the times it can just cause permanent neurodegenerative like neuro effects basically their development is completely off from the get-go right yeah and so very quickly um you know the the good doctors at a chop they figured out that i was dr rebecca aaron's nicholas i hope i get that right but You know, she is a pediatrician at CHOP. She's working with this other cardiologist, Kieran Musonuru. We'll talk about Kieran in a second because he actually has a pretty fascinating career. But... they kind of were thinking, you know, we can do something for KJ that hasn't really been done before.
We know exactly what his DNA, where it's broken. We know exactly what organ we need to target. And we have parents that are really bought into this idea of like, let's try something radically different for him. And they knew which DNA was broken because they had identified the particular condition.
Speaker 1
And this is something that had been studied before. And so it's like, we know what we're looking for.
Speaker 3
Yeah, exactly. Yeah. So, you know, as I was getting to, like the cycle's known. Like we know exactly what should go to what. Like this goes to there, that goes to there, and it's a loop. And so when you have buildup in one thing, you can just sort of determine, well, it's probably that enzyme that's broken. Like if X is supposed to turn into Y, but it doesn't, you have zero Y and like a ton of X, you know that the thing that makes X into Y is broken. So that was the case for KJ. They sequence this DNA. They figure out these are the mutations that are causing him to have this. And fortunately for KJ, the story of how this came to be for him started a couple of years prior.
The therapy that they ended up developing for him was something that was unique for him. But there was a blueprint that Kieran and Becca had sort of built out with some other collaborators. over the last couple of years. And maybe like a good place to start is we can sort of talk about the elements involved and kind of how that plays out. But ultimately what they did for him was they figured out his mutation and then they say, well, we kind of know how to go about the process of making something specific for KJ. So he ultimately is the first baby born, the first human who had this specific therapy made for him.
So there've been a few, there've been a number of therapy trials. KJ was like the 18th, we'll call it trial, right? But it was just one patient enrolled in this trial. There've been other trials and there are other gene therapy products on the market, but those are all kind of marketed at mutations that are like, conserved within certain gene processes. Like sickle cell disease kind of has, you know, many people get sickle cell, but they all kind of have the same genotype, the same thing that's broken. It's a bit broader. Yeah. KJ has this disease that's super rare. And, you know, there might be dozens of children born with that.
It's like a really small market. And they all might have like very subtly different mutations that are driving. So the gene is broken, but the thing that actually specifically is broken within that gene might differ. And to set the stakes, you hit on some of this, but the clock is ticking, right?
Speaker 1
Because he's got... developmental delays that are severe and and this ammonia buildup is quite dramatic and and so that you know everybody's is upset and freaking out and and then like you said i mean in the case of like sickle cell i mean this is something that's being studied over a long period of time and you're trying different things and people are going to be at different stages of life but with this particular baby it's like oh we gotta we have to address the situation and you can do some therapies to kind of lessen the amount of ammonia but i mean it's it's it's like it's pretty when i was reading the story i mean it's it's like it's dangerous stuff and and and yeah i mean it's you you want to take action and figure this out so he's in this lucky position where why why had people been thinking about this particular thing two years prior
Speaker 3
Yeah, so the story that I think is actually quite fascinating. So obviously what was done for KJ is remarkable, but there's a little bit more that happened before that. So they were basically ready to go by the time he was born. what Kieran and Becca and collaborators like David Lu, I think Doudna's, IGI in Berkeley, there are some other folks there, Petra Stanakopoulos. So a bunch of people are kind of in the mix. There are a few companies involved. And what they had been trying to do for a couple of years is get down this feedback loop. And if we can just touch on something very briefly, I think in biology, it's very different than computer science, very different than AI.
In AI, you have an idea, And you kind of like hack on it and you can get a sense for whether it's a good idea or not, whether it's possible very, very quickly. Biology is very different. It can take, you know, if you're lucky, it could take you weeks. If you're working with standard biology, it could take you months or years. And in the case of what they were trying to do, they were basically trying to say, you know, for any given mutation that we're interested in, so like urea cycle disorders is what they were looking at. So any of those enzymes I was alluding to, they were saying like, can we make a blueprint to go to and say like, okay, baby's born CPS1, this is what we're going to do to get a therapy for that baby.
So, you know, we're going to sequence and then we're going to want to be able to sort of replicate the specific mutation in a cell line that is specific to that baby that's born. So in this case with KJ, what they do is like immediately they're sequencing his DNA. And then there's a group at the Bird Institute led by David Liu, who's a really important figure in protein engineering. But I think these days, uh, serious gene editing type stuff. Like it's, it's, you know, his lab is, is really important in that, in that, uh, in that sphere. And so, you know, David Lou's lab is sort of lending a hand where Kieran is coming to them and saying, like, what kind of editors can work?
There are a lot of options on the table. And, uh, and then there is the idea of like, well, how do we deliver that editing? So then they're working with, um, uh, industry to find exactly like what options are on the table. And we can talk more generally about that in a second. And then you have a group at IGI in Berkeley that is sort of figuring out like, is this therapy safe? And so what the loop looks like for them is, baby's born, figure out the mutation, make a cell line and a mouse model specific to that mutation for that baby, right? And so you kind of have like KJ liver cells, not taken from KJ, but like regular liver cells that folks in like lab biology work on.
So you grow those, you kind of make them look like KJs, DNA that's broken. And then you have a mouse model that's reflective of his thing as well. And the idea there is we can test very quickly whether the therapy that we're designing is actually fixing the mutation. And if we can just step back very quickly briefly, or maybe it won't be so brief, I don't know. I can elaborate a lot, but you catch me if I'm going too far. But I think there's an important distinction here to make with gene therapies that are very different than the modalities of therapy that have existed before. So gene and cell therapy's idea is like, You know, we have cells and they break in certain ways.
And for a long time, the way that we went about fixing those things as we go, like, well, we'll just put like magic goop, like a chemical, and that chemical will like do enough stuff to like kind of fix the end phenotype. So, you know, you have breast cancer and we give you chemotherapy and maybe we know how the chemotherapy works in advance. So like in 1960, maybe they didn't know. So they would just give you a chemical and it works. It's a small molecule generally. And it can do a lot of stuff because a small molecule interacts with a lot of things in a cell. Gene therapy is a little bit different, right?
So you're not kind of like putting a chemical in and sort of trying to observe what the effects are. You're kind of going in with a very precise idea of what you want to do. It's a sort of mechanism guided design. You know exactly where you want to go and you know exactly what you want to fix. And it's an important distinction because that's, I think, a big part of the story for why this was possible for KJ. You can develop a sense for what you want to do. And then design around that. Whereas with like small molecules, you say like, I want to kill cancer cells and like, you just hope you get lucky and you find something that can do that.
Now there's more complication these days in terms of what people try for that. But historically that's been the case. And nowadays with gene therapies, with these rare diseases, uh, For KJ, without a gene therapy, the options were enzyme replacement and supplementation for the metabolites that are missing in that loop cycle for him. He's taking some kind of medicine forever. Yeah, forever. It's forever. In this case, the hope is we can just return the function that was lost.
Speaker 1
There's no way everyone listening to this gets us the like when you guys talk about a gene therapy like you're describing i mean is it going in and rewriting the code the dna in every single one of your cells or you're triggering some cells to produce this enzyme and you just need enough of them to kind of make what are you actually yeah doing yeah that's a yeah uh all right i'm gonna i'm gonna do it i say all right ashley let's let's let's
Speaker 3
step back let's step back Eventually you're going to have to pull me into like whatever direction you want to go.
Speaker 1
It's inception. It's gene therapy inception right now.
Speaker 3
Let's set a framework together for like what is a gene therapy. So at the highest level, the idea is you have DNA you want to fix. And the way to think about designing a gene therapy is you have like a few questions in mind. The first one is like, what organ do I want to go to? Like what organ is the problem? Another question is what gene is broken or like what do I need to fix? And then the third one is like, do I want this to be for a moment? Do I want to fix it sort of short term? Do I want to fix this like forever? And those are kind of the things people think about, right? So we have our framework. I'm sort of borrowing the three bullet point. I know from Sarah, Sarah Payne. I don't know if you've been watching like the Turkish, but she talks about like three point framework. So we'll do that. We have three points we want to hit. Where do you want to go? Where do you want to fix? And how long do you want to fix it for? Bro, if you get me Sarah Payne numbers of this pie, we're coming back every week.
Speaker 1
Yeah, yeah, yeah. I don't know if I could help you with that one. Okay, since day one, the Core Memory podcast has been supported by the fine people at E1 Ventures. They are a young and ambitious VC firm in Silicon Valley, investing in young and ambitious companies and people. Thank you so much to E1 Ventures for all your support.
Speaker 3
Okay, so we have our framework, right? All right, so now we have three things we want to do. Now, how do we actually do that? So let's design a gene therapy for CPS1 deficiency. So that's exactly what KJ has. Well, we think about what is broken in KJ's, like what organ is actually the one, the problem area? It's the liver. Okay. And in his case, the liver is the place where this loop cycle is going. And so we know we have to go to the liver. So that's our delivery thing. We're going to talk about delivery in a second. And then we know what the gene is, CPS1, easy. And we know what the thing is that we need to fix.
And the last one is, do we have to do this forever? So in his case, the answer is yes. uh and with gene therapy so now we're like we're setting our frameworks um and then the last part to talk about is like how do i actually go about doing that so you have your your payload your payload is like the stuff that's going to do the action in a cell so that's the crispr cas9 historically people talk about that in the last decade we have this like magic sort of enzyme that can go in and fix at the DNA level something. And you have different options for things. So like there's the original CRISPR-Cas9 system as it was envisioned for use by Berkeley, Broad folks.
And then you have sort of the second generation of things, which is the base editor. And the base editor is the thing that ended up being used in KJ's case. And then you have these new generations of things that have come after the base editor of the prime editor. There's stuff that's happening at the ARC Institute as well, RNA bridge. So there have been things afterwards, but there are different mechanisms that those things do. But OK, so that's that. But you have your cargo. then you have to have a thing that guides that enzyme to the specific gene site that you want to edit. So we call that guide RNA.
And that's the thing that's sort of homing the enzyme to the site you want to work within. And then, okay, all that is great. It's all wonderful. But you need to get it to the place you want to go. And the last part that has gotten a lot of activity recently and you know i worked on this for a long time at dyno it's like delivery gene delivery so it's not enough that you can edit things in a dish right you have to be able to edit in a human and there are a lot of intelligent things that doctors and scientists have drummed up but in kj's case they end up using a lipid nanoparticle sort of like a fat it's like a fancy way of saying like a fat uh droplet that is containing these enzymes.
It's mRNA. We won't complicate it, but basically fat is delivering this enzyme. It's delivering the guide to get it there. And there are other options on the table which they thought about, but depending on where you want to go, you either want to go with a lipid nanoparticle, you want to go with a virus, you want to go with a lot of different things, and there are many options on the table. But if we return to Cage's case, right, so we have our delivery mechanism, which is a lipid nanoparticle. It goes very, very well to the liver. We have our editor. So in Cage's case, there's one, he has a couple of mutations, but they decided we're gonna fix one specific thing.
We wanna turn an A, an adenine, into a G. So we wanna go to the DNA level, and the cps1 gene and turn one base pair into another base in these liver cells in the liver cells right and so and then you have a guide that gets you there and towards back to your question i promise we took a walk and i'm coming back i want to learn this yeah yeah great we're coming back to so to your question like how how does that actually work so in an ideal world you can edit a hundred percent of the cells that you want to edit In the real world, we're not quite there yet. So, you know, I can't, I don't know for certain what percent was in Cage's case, because ultimately we can't know these things.
We can have guesses, and that's what the animal models were for, was like, can we come up with like a rough sense for like, how efficient is this? How many cells are we actually able to target? And I think they were, we'll sort of ballpark at like 40, I think it was like 42% is like where they arrived at with their cell and mouse model work. And turns out, and you know, it's an open question, like, is that enough? We don't know. And in cages, it's just- To fix, like, you could be having some effect, but is it enough effect? Yeah, exactly. Like, there's probably some threshold, but after which, like, there are enough functional cells that the body can just, like, make it work.
And, you know, maybe they didn't know in advance, like, Is it going to be enough? And in his case, it ended up being enough to at least reduce the amount of the typical medicine he would have received. They just slashed that by 50%. And there are open questions about, well, you have to get another dose. It's possible. But there are a lot of questions that have been actually answered by the study that were not really known before. And a really big one, and I think a really cool one, is we learned you can actually dose somebody with the same therapy more than one time. And you don't have to worry about immune effects, which has always been like the big elephant in the room with gene therapies.
Speaker 1
And maybe you're getting there, but just to be super explicit, once you've done everything you were describing, I mean, the liver, like in my pedestrian head, I'm picturing this liver with millions upon millions of cells. It's already, it's kind of, it's got its identity. And then I'm picturing this thing going in and like, rewiring each one of those cells or are you just telling the liver like from now on you know we want you to to produce this certain thing and to go this like are you yeah are you kind of taking it all these existing cells and fixing them or this is just moving forward this is what i'm trying to understand is what actually is happening in the liver cells
Speaker 3
Yeah, it's an insightful question. So I think the core principle to sort of go to is the cells that we have are propagated forward. So you have liver cells today, and they divide, and they will be the sort of parental generation of your liver cells in 30 years. Right. Well, in your, I think 50, maybe 100 years in your case, right? So they'll be around, you know, if Brian Johnson has his way and, you know. I drained his blood when I was here. Oh, is that right? Yeah, okay. So, you know, you kind of, you go to like the cells that exist. And you fix them with the knowledge that they're going to divide eventually, and they'll be the next generation. So you go to the cells, you know, we'll call it the generation, the now generation. You fix them, and then they're going to divide. And now the new cells that come about later will be functionally repaired. Yeah, okay. Yes.
Speaker 1
You're sending these instructions. Mm-hmm. Yeah.
Speaker 3
Yeah, yeah. Instructions for moving forward. Exactly, yeah. And, you know, again, do you have to fix 100%? It's unclear. It's probably disease process. Some diseases you will, some diseases you won't. And in this case, it turns out you didn't need to be 100% to at least make an effective reduction in the amount of medication he needs. And that's a pretty huge thing. And, you know, it's an open question about whether... He will eventually just be weaned off of it entirely if he maybe gets another dose. And all of a sudden it's like, oh, we got to 70% or 80%. And now all of a sudden he doesn't need medication. And we learned enough, at least in this case, to know we're comfortable. We can say, well, we can give him another dose later on if he needs it.
Speaker 1
So you had these various groups who... This is a question they were already interested in. This is the work they specialize in. And so they had been trying to develop this framework over a couple of years of, if something like this pops up, how do we try to act quickly? You've got all sorts of problems, right? I mean, it's high risk. It's vaguely controversial because of the risk. And so then you have to deal with regulators in the midst of trying to figure out a therapy for this child who's in distress. And so... So I mean, I love your story because part of it is this backstory of how all this came to be.
So you've got people doing this work. They have to figure out almost like a general purpose mechanism, which then you can slot in. You need a general purpose, like approach baseline. And then you need to be able to prove you know what you're doing to somebody for when a crucial moment comes. And then so they've been doing this work for a couple of years. And then baby KJ is born. and then the bits that you're talking about with the the mouse models and everything you also have to be able to spring into action quickly to like adjust to this personalized yeah therapy but the a big heart of your story was this all of this work that went into this over the couple of years to prepare for the situation
Speaker 3
yeah yeah exactly yeah so you know um So what exactly were they working on for two years? They were basically just like, they were demoing. It's sort of like drills, right? Like, what do you do? You play tennis, right? I hear you're pretty good. So you play tennis, and you're not playing... You're not playing, you know, a full, you're not, you're not doing a whole thing every day. Right. But you, you run drills, like there's certain tournament or something.
Speaker 1
Yeah.
Speaker 3
You're not doing that every week.
Speaker 1
Right.
Speaker 3
But you, you sort of practice things you want to work on. So in their case, they, they, they might've run a drill where they're saying like, we want to work on OTC or a thin transcarboxy. So like another enzyme that's broken in that pathway. how would we go about making a sort of pathway to fixing that in a human? So you sort of dream up the mutation, you make a process for picking an editor, you make a process for picking a delivery vehicle, you make a process for making cells that are specific to that mutation, and you come up with a deal, or maybe deals, but you come up with an understanding with the FDA about what they would like to see to know this is going to be safe and it's going to be effective.
And, you know, I think an important thing to come back to, you're talking about safety, you're talking about, like, there's an ethical thing. The good news in KJ's case is, like, all the parts here were proven out. Not even in this loop, right? It's been proven out by other groups. So LNPs, lipid nanoparticles, they've been proven out several billion times. It's actually the thing that was delivering the COVID vaccine to... Many, many, many people. So we know like, yeah, there might be some side effects, but largely like it works. It works extremely well at getting to certain types of tissues.
The base editor had been trialed in other trials as well. So we know it can work in hemophilia. We know it can work in sickle cell, in human beings. So like that element is working. The LNP is working. And we know we can design guides very well and check for offsite mutations because there's been a ton of academic and industry literature that's been generated on that. So KJ's born, the group is sort of working off of information that's been gathered in the field over the last five years or so. But also they have been building their own loop and a process for like, oh, we got a new mutation. How do we make this work?
And they really kind of like, it was very clinical the way that they went about it. It was like, it was not a mad rush. It was very much, oh, we got this mutation. Okay. There was a grad student actually in Karen's lab, Sarah Grandinette, who did a lot of the the lab work. So imagine that for like PhD thesis, you can see like, oh yeah, by the way, like I fixed this like gene therapy. No one's ever done this thing before. That is a pretty cool thesis. And so, you know, they have this process, The gene's broken. Okay, let's make some cells. Let's make the mouse models. Let's talk to our industry partners.
They can spin up the LNPs. They can spin up the production of the mRNA. They can spin up a lot of things. The Broad Institute or the IGI can help us pick an editor and do some simulations for whether it's going to create off-site mutations, which is kind of a concern with these therapies because you're permanently fixing DNA. You just want to make sure you're going exactly where you want to go. And so all this kind of, it took some time, but what they were doing for two years was trialing with other types of mutations. And KJ just happened to be a baby that was born with CPS-1 and they were like, yeah, I think we can make this work.
Speaker 1
Right, yeah. And so from the time they identify his condition to when he gets a therapy, how much time passes? It was five months.
Speaker 3
Five months. That's like five months. That's insane. That is insane. I'll just say like, The typical clinical pipeline for, you know, for instance, the sickle cell or beta thalassemia gene therapies, it was like a decade, 10 years. This was five months and it was safe and it was effective. And they did it with like an understanding with the FDA of what they needed to see. But it was extremely rapid. And I think there's a lesson there.
Speaker 1
Yeah. Okay. And then just to compare this to, I mean, you mentioned we have had gene therapies. They're rare. They're very expensive in different sort of situations. There was the Chinese doctor who did super controversially editing on... before the children were born to, if I remember right, to prevent them from getting HIV and was thrown in jail for three years for doing that. The difference, a couple of differences, I guess, but, you know, they did not have a life-threatening condition. He was just altering these children. Obviously, this was years ago, so technology is less mature as well. Um, but that was super controversial. But when we come to this case where you're taking a risk on, on this young child, but this child was going to suffer. Right. And if you, if you didn't do something, I mean, is that, is that sort of some of the ethical differences here and why, why this was such a celebrated case versus him being imprisoned for, for three years?
Speaker 3
Yeah. So, you know, there, there, there, a lot there. Yeah. There are a couple of distinctions I'll make. So the first one is like Kieran Misunuru, Rebecca Arons-Nicholas, David Liu, Fyodor Urnov, Petrosina Kalbos. They're not John Cui. These are very serious people who were going about doing it the right way. The problem that I think John Kui ran into, the Chinese scientist, the problem he ran into is, as you were sort of getting to there, he wasn't fixing a disease. I think my understanding of things is like the father of the parental DNA, The father had HIV, I think. And it was a sort of like fundamental misunderstanding of like, well, the children will have HIV because the sperm has HIV.
But it's actually not the way that HIV works. And that was at least his like veil to do the work was like, well, the father has HIV. HIV can be given to a baby if the mother has HIV. But we actually have medications that can prevent that. So ultimately, this was like, one, a non-issue. And two, he was editing the embryos. So these were not babies that were born that were then edited because they had HIV. These were embryos that he was like, well, like, we'll just, you know, we'll just do stuff to them. And there are a lot of technical problems that are involved. There are a lot of ethical problems that are kind of related to, you know, the fact that he's not really fixing something.
And because of that, it's kind of unclear what he's really adding to the value of their life. And also, you know, maybe the parents didn't have a firm understanding exactly what was going on, like what the risks were. In KJ's case, everything was very crystal clear. The parents were told, you know, very plainly exactly what was happening. It was done extremely with, you know, with all ethical considerations in mind. And it was fixing a real disease that was going to likely, I mean, 50% of babies born with this condition die. It was likely going to end up in a bad outcome. So, you know,
Speaker 1
No, very different. I just wanted to cover it because it was like these two big famous moments, infamous famous moments. And I just wanted to, if people had kind of caught news of this in the background, just to give them some context around the differences. And we should say Baby KJ is doing quite well.
Speaker 3
Yeah, yeah. So, you know, they did you know, they gave baby cages born in November. They design all this stuff. They ended up giving him a therapy, I think in February, March, he actually ended up getting two doses. The first dose that they thought would be enough was like too low. So they dosed him again and that was enough to do what they wanted to do. And it's been, I mean, it's been a while since, you know, the paper came out, but they had actually tracked him for a couple of months after birth to see how he was doing. So markers of improvement there, like, can he eat protein at the level of like a typical baby?
good. Yeah, he could do that. And he wasn't getting buildup of ammonia in his blood. He also was able to fight off infection. So when we get sick, it causes a lot of stress, metabolic stress, and this can lead to increase in ammonia. If your body is able to do it fine, you have all the enzymes that are working. It's a no big deal for the average person. But in his case, without the therapy, it probably would have been a struggle for him. He would have had a spike in the ammonia and then You kind of got to work with the medications if they're an option. But we know that it was sufficient because he fought them off and he wasn't dealing with these crises where you have buildup of ammonia from those infections.
So yeah, by all accounts, he's doing quite well, yeah. And you get to this at the end of your story, but what does this mean for the world now?
Speaker 1
Does this mean... Um, we have similar things in store already for a large number of these rare genomic diseases. Most, a lot of them are monogenetic. So they affect a single gene. What about the multiple genes? What does baby KJ represents for the next like five or 10 years? Yeah, so I think there's the acute thing, which is there are two big lessons.
Speaker 3
To me, two big lessons. So the first one is we actually learned, I think, that all the tools necessary to edit monogenetic diseases are available to us. We have plenty of functional delivery vehicles. We have... a lot of safe editors to go to and we have ways of designing guide RNAs to effectively deliver all that stuff to exactly where you want to go and ways to check that you didn't make a mistake so we learned that you can actually do this for a lot of single gene aberrant mutations like we can fix those I think and what we do with that information i think is ultimately up to you know the powers that be right so if you're working on a rare disease i think you could probably have a little bit more confidence that one we could do it and then two this is the second big lesson uh is that the fda is like you know actually we're willing to like refine our process to work with the understanding that this type of therapy is like a little different than like the thing we've been, the mode that we've been built on, which is like small molecules.
They're willing to like play ball a little bit. And I think that is a different and very important type of lesson. So that's like the short term is like, I actually think My guess is we're going to see a lot of therapies like this. Uh, you know, if you do them safely and you do them in the right way, as they did in this case, you can really do quite a miraculous outcome. You can get a really outsized return. Uh, and if you pick the right, if you get lucky and your problem set is like in an area where you can use like lipid nanoparticles, which are very, very cheap to produce, like a dose of an LNP might be like a few bucks.
Um, Like, you can have a cheap, effective gene therapy ready to go. And then the FDA might be like, show us this type of data and, like, we'll work with you. There are still things that need to be revisited, I think, in terms of regulation. But, you know, we can touch on that in a second. Thinking about the future... It's a little bit harder. So, you know, I was talking to David Liu about his thoughts. Like, what does this mean? What types of diseases are on the table? And here is like a pretty important distinction, which is, you know, we can edit single genes very well. With more complicated diseases like maybe Alzheimer's, Parkinson's, type 1 diabetes, there are a lot of diseases where it's a little bit more opaque.
We're not really clear exactly what drives those processes. If we revisit our framework of where we want to go, so in Alzheimer's you're like, well, we want to go to the brain. I think that's fair. Sure. Probably. And then the second question is like, oh, do we know what we want to edit? We run into problems, right? We actually don't know what we want to edit. And so if we think about the future, the biology is actually the gating factor, like the understanding of the diseases, it's gating things. And there are a lot of diseases. So, you know, rare diseases are generally like single cause. Like there are things that are driving that we can understand very well.
And historically, the thing that stopped us is that we didn't have the tools to address it, but now we do, great. But then these more common diseases, maybe there's a lot more at play. There might be dozens, hundreds, thousands. We really don't know in some of these cases. And is there a single weight-bearing gene that's breaking the thing? Maybe. So in Alzheimer's case, not to cloud things too much, but there's been data that's shown that certain genotypes for certain fat genes uptake and like there are certain uh proteins that you know deal with fat so apo uh e there are two three four people who have a certain phenotype which is apo e4 so two copies of that develop alzheimer's like 100 of the time And can we say that that's causative for Alzheimer's?
No, but maybe their ideas here were like, maybe you can actually preempt the development of Alzheimer's by fixing that early on. So maybe when someone's 30 or 40, you know, you sequence them, you figure out this is the gene that's broken and maybe you won't guarantee they won't get Alzheimer's, but instead of getting it at 70, they might get it at 85. And I think that's kind of magic. But then there are a lot of diseases where we don't really have any case. We don't know. ALS, like, unfortunately, we really, there's 5% of people have a familial genotype. SOD1, I think, is the gene that's broken.
But in the 95% of cases where that's not an option to fix what do you do and so to really like go back to your question what is on the table in terms of more complex diseases unfortunately i have to say uh the biologist needs to be the biology needs to be worked out a little bit and luckily we have groups like chan zuckerberg institute you know they're doing really they have a moonshot that i think is hopefully going to be like, that's the sort of stuff that we need to like break open our understanding. And we can be hopeful, you know, we're in Palo Alto, so I think I can be hopeful and say like, maybe AGI will like fix all this stuff and maybe it'll make us like understand biology differently.
But until that point, we kind of, gene therapy is kind of a tougher pick for those diseases.
Speaker 1
I was going to ask you some of the Silicon Valley questions. You can check out our episode with Priscilla Chan to learn about the Chan Zuckerberg Institute. I have so many questions for you. I'm looking at the time. Can I veer off, or do we have a couple more notes on that?
Speaker 3
Yeah, no, you drop the ship here.
Speaker 1
Okay, okay. I mean, I've always been a generalist as a reporter, so I'm not like a biotech whiz, but biotech has always been one of my favorite areas to go into. I just don't want to embarrass myself too much. But, you know, I remember going back about 15... And this was pre CRISPR and it was, it was like the DIY bio movement was happening. Drew Andy over here at Stanford was a major figure in this. And, you know, I remember going to my futuristic conferences and everyone's like, biology is now just computation and we're going to create enzymes that. we're going to grow chairs in a, in a factory out of a vat, like, like, you know, beer, we're just going to make chairs and cars and, and we've figured out DNA and, and all these mechanisms.
And obviously that did not happen. Um, and then, you know, that, which I'm, I, It's like, I love these ideas and I'm not being like crazy critical in some ways because all these things take time. I mean, even when we sequence the genome, a lot of people are like, oh, the next day, all disease will be solved. Now that we know this code and we've seen it's taken 25 years of very hard work to like make progress on some of these things. So I get that. But, you know, CRISPR came after I was already hearing all this stuff. And so it was like, people were already excited. Now we have like, apparently this miraculous editor that can do anything.
And then, like you said, about 10 years have passed. And when I go report on some of these stories with startups, it's like, oh, well, we still have problems with like long strings of DNA. And then it turns out like, yeah, we can make the thing in the dish, like you said, but organisms aren't, things go wrong when we start making these cuts and all this stuff. And then you just laid out this whole concrete story where this happened. But I still feel like I am at a loss of how excited to get about CRISPR because I seem to run into, there's a lot of groups trying to make these longer, longer cuts in DNA and, And then the delivery mechanism of how you make that work.
And when I go interview these startups, I mean, they're getting funded with hundreds of millions of dollars. Some of them are going under. This seems like a big problem. So I can never tell. I was told CRISPR was like a panacea for fixing all this stuff. As always happens, these things are harder. But given all that context you just gave us, I mean, what stage are we at with this as a super fantastic editor for DNA?
Speaker 3
Yeah. Well, I think we can open the answer to your question by, you know, what I'll say is this. Biotech is done for the love of the game, Ashley. It's done for the love of the game, right? So, you know, what does that mean? It's actually, it's a very tricky business. So... I think about like prime medicine. So it's David Lu, sort of the technology that came out of his lab, the prime editor. It's like, you know, a new generation of editor can do these huge edits. So as you're getting to like, we can edit much bigger space now. And they're showing that it's working very well. Actually, they just had a trial.
It's ongoing trial, but it's going very well. But Prime Medicine is not NVIDIA. It's like the stock is not printing cash. And I think ultimately, there are a lot more intelligent people than me who can talk about this. But I think it's a tough business to get into. So there are many ways that I could answer your question. But I think ultimately, what I go back to is you need to have a pretty strong stomach to do biology. I'll sort of borrow from Lee Kuan Yew here, which is like whoever does biology, you must have iron in you, right? You have to be willing to deal with a ton of uncertainty. And so that's like the editor stuff.
Which is, unfortunately, the stuff is going very well. The actual mechanics are excellent. We can deliver to the liver, as we learned with L&Ps. I worked at a company called Dyno Therapeutics, and what we worked on was using machine learning to design novel capsids, novel viral-based delivery vehicles that can go to more specialty areas, so not just the liver, but deep regions of the brain, the eye, muscle, like things that LNPs today can't do. And then there are people working on other modalities, but, and then the editors, as we talked about, are like working, but the business is tough because ultimately like making a gene therapy is absurdly expensive.
And that really comes down to, and the hope is economies of scale will solve this problem, but ultimately it's very expensive to make a viral vector based gene therapy. the, the reagents alone will cost you for a human. So for, for an adult, so we, you know, let's talk about a kid actually, how much might that cost if you use AAV? Well, we learned with some of the like Luxterna, which is like, you know, gene therapy for, um, for, uh, a retinal disease, uh, it costs like millions of dollars. That's, that's how much it costs. And you know, how much of that is the cost of running trials baked into the thing is, is different.
a sort of different thing to address, but the actual material that goes into the therapy is very expensive. Just the chemicals. Yeah, so you have to do it in a very safe way. You don't want to have contaminated reagents. Ultimately, this is like raw stuff going into a human body, right? We don't want to activate the immune system to make someone very sick, independent of the therapy. So people do this safely. And the good news is that, largely speaking, our process is quite good. But the plastics are very expensive. The media is very expensive. The labor is very expensive. And that's because the biology of how we make AAV is...
Kind of requires like one hand to one, you know, hyperflask and you got a, you know, gene therapy for the eyes made one hyperflask at a time. And those things are very expensive.
Speaker 1
I thought we just ordered our reagents off Alibaba from China and they... show up yeah uh maybe maybe that's the future but we're not there yet uh you know they still unfortunately have to deal with the same like gating problems that in america we have to deal with and i'm hopeful we can figure it out but we haven't figured it out and ultimately these things are expensive and so are people going to those therapies when they're available it's kind of not as clear i think right now which is unfortunate because they they you know in a lot of cases they work very
Speaker 3
well um And I think the hope is as more become available, the cost comes down and we become better at identifying what are the problem areas that make this expensive. So that's the reagents. Then there's this entire thing about the FDA process of review that makes making a therapy kind of a scary thing. So if you're We can talk about not gene therapy in a second. I think there was more to your question about biotech, DIY bio more generally. But talking about gene therapy, there's this element of regulation. So as I've been hopefully getting across, gene therapy is a very precise thing. You can know a lot before you even do it about how things are going to go.
You can know if you're going to edit the right place. You're going to know if you're going to have off targets. You can actually know a lot of stuff going in. The question, I think, is more, you know, we have a gene therapy today for CPS1, turns out. LNP, there's a certain one, you know, that they use. And then there's a certain base editor that came out of David Liu's lab in 2020 that they ended up going to, an adenine base editor. So it swaps one mutated amino acid, so like adenine to guanine. So A to G, it fixes that. But it's the 2020, you know, invention. And then the guides rate. That's unlikely to change.
But those first two things might change quite a bit. So let's just say in 10 years, there is a much better editor. It's just so much better. And it's been proven out 1,000 times in other therapies. And you want to fix CPS1 still. And let's just say the LNP is the same, but now we want to use a new adenine-based editor that has a higher efficiency, lower off targets, whatever. Like, do you have to do a whole new trial to do that? And I think that is the sort of question that the FDA, I think, probably has to wrestle with. And, like, hopefully the industry can contribute to the conversation to drive us in the right direction.
But I think there are a lot of advantages to gene therapies that we can work with to refine the process so you don't need to run these, like, highly expensive trials that just tack on to the cost. How cheap can we ultimately make a gene therapy? I'm hopeful that you could make a one-time gene therapy for, I'm going to say a scary number. I think you could get to low hundreds of thousands of dollars. Is that an absurd number? I don't think so. Because a lot of monoclonal antibody-based therapies that we use for a lot of diseases, rheumatoid arthritis, Crohn's disease, very common diseases, people pay...
you know, tens of thousands of dollars a year. And they do it every year until, you know, they no longer decide they want to do it anymore. That might be 20 years at $20,000 a year, you're paying 400K over a lifetime. What if the gene therapy is a one-time thing and you do it and it costs 200K? It's not crazy. It's not a crazy number. But you need to find a way of making it palatable um agreement with you know payers and it has to work out but ultimately like i think the cost problem needs to be addressed and that can be done in a lot of different ways make your process cheaper make your regulatory framework a little bit more pliable safely but pliable and And you can drive down the cost.
And I think that will open up a lot of appetite for life-saving therapies. Because as you started out at the top of the show, you said something which I think is important, which is we would love to fix a lot of diseases. And it turns out we can. But it's very expensive. And if people aren't going to buy it, the incentive to work on that problem is pretty low. So that's gene therapy. Now, I don't know. Do you want to talk about DIY bio more generally?
Speaker 1
I do. And just super quick before we do that. I mean, how does this compare to where is the race between US and China on this stuff? I mean, you could see a scenario where the government there Um, I don't actually know. I mean, maybe their version of the FDA is more flexible on some of these things. Maybe the government is more willing to fund some of this stuff as kind of like, because they are a, it's like communist society. They take care of their people in different ways than the U S has chosen to do. Um, and so, um, you know, are they are, yeah, where, where are we, uh, in terms of, of the cutting edge of, of this type of work?
Speaker 3
Yeah. So, you know, if we look at the number of trials that have happened for gene therapy, I think there have been 19 trials now. There are like 17 base editors and now like two prime editing trials, or maybe it's 18 and one. But there have been a number of gene therapy trials. I think about maybe like a third to half have been done in China. China is like, we've awakened the sleeping giant of the biotech in China. They're now like a fairly robust industry there, and they're doing things that we can't do in America. It's There are still open things to work out. So things that I don't think are quite clear yet are, can China work on new modalities that don't exist in America?
I don't know. I don't have an intelligent take on that. However, what we have learned is that China is willing to take academic work that's been done in America and just very rapidly deploy it to the clinic much faster than America can. years faster. And that is a pretty bad calculus if you're, you know, if you're an academic and you have an idea and you're thinking about, like, I would like this to be in the world. Do I want it to be, if I want it to be in the world, like, do I go to China or do I go to the US? And I think I would love for it to be in the US. Like, that's my dream. Like, I think we should reap the benefits of the American, American taxpayers investing in science, right?
And we've done a lot of the legwork. we shouldn't be kind of giving it away to China, which is unfortunately how it's worked out. They have a lot of, you know, for better or worse, like their regulatory frameworks are, they're a lot shorter, like their things are willing to do that we don't do in the U.S. because we're very safety oriented, which is great. But they also can get ahead of us very quickly. And so It's not clear if it's happened yet. I don't think it quite has. But there has been momentum shifting in the direction of China. Like the balance used to be like this. The U.S. is like, you know, way, you know, they were like really ahead.
And now it's like we're kind of at 50-50. I really don't know how it could go.
Speaker 1
And this may be off too far off to the side for this type of work. Maybe not. I mean, also just in terms of doing non-human primate, um, studies, I mean, China, China has lots of non, I mean, it's, it's a difficult thing to talk about. It's an unpleasant thing to talk about. Um, they have, they got lots of monkeys to do all kinds of experiments and we, we, we don't. And, and, um, yeah.
Speaker 3
Yeah. They're, you know, I, I would align personally more with the American model where you like really, you, you take very seriously the charge of being like a human, which is like, you don't, you don't take advantage of like the American, the human like intelligence gap over other animals. You don't just kind of like tort, you know, torture is not the right word, but you don't like run experiments just because you can. Yeah. You really think quite strongly about what you're doing. And in China, I think they're more willing to just, you know, like, let's, let's see if this works. And I think, It's not to say they're doing things that I think are, like, wrong, but they're certainly willing to push a little bit further on, you know, push a little bit more on animal studies than, like, we do in the US.
There are a lot of things that you need to do before you go into a monkey. Culturally, you know, it's not like the FDA says, like, you need to do XYZ before you go into a monkey. It's just that's the way we think about it in the US. The monkey is like the step before the human. And so we take that very seriously. In China, it's like, well, why do it in a pig? You just go straight to the monkey.
Speaker 1
That's the thing, right? Because the mice, the pigs, all this stuff have so many limitations. And it's like, well, if we have a... Yeah, something a lot closer. Let's find out. I do want to talk about DIY bio. Going back to my long-winded setup before, I just remember it being like the most exciting thing. It was like so cool. It was beyond computing. I feel like Josie Zaner, who's a friend of mine, they... are always on the cutting edge, sometimes controversially, but seem to be at the vanguard. And a lot of people that used to talk about this, it's like not in the zeitgeist. It's not buzzy.
Like, am I right? I mean, I remember even down the street, all over Silicon Valley, there used to be DIY bio clubs. Those have gone away. Clearly, we have biotech startups doing adventurous, interesting things, but this kind of like... um maker movement of biology i feel like at least maybe i'm just missing it it just doesn't feel the same to me yeah i yeah i mean i think you you're you're you're right there's not you know if i think about what kinds of companies exist in biology and i like think about how that money is deployed ultimately that's probably like a pretty good estimate of like where the company interests are i think pharma is probably close like everyone's worrying pharma stuff net pro like 95 plus percent is on that because you know if That's probably what the average biologist is like, yeah, I want to help people, right?
Speaker 3
I want to do that. And if you're VC, it's something you can rest your hat on. You can be happy with that and investing, deploying capital to solving those problems. Consumer bio is difficult because what are the interesting things to work on? So I have a personal... I'm very interested in like plant engineering. There are a few people that I can name that, you know, Nick Desneuer, who does like these fabulous, fabulous, like amazing artistic displays with flowers. Asimov Press just did like a really great interview with him. Where's he based? I think he's in Switzerland. Okay. Yeah. I think he's in Switzerland. I'm 70% on that. Sounds like someone I need to go visit.
Speaker 1
Yeah. Yeah. I know. Yeah.
Speaker 3
We can talk. Yeah. But so he does like amazing painting. plant displays and then there's a company you actually buy from light bio they make i don't know if you saw these like glow in the dark oh yeah yeah yeah so like you know do these things exist like yeah and maybe there are other things that we don't consider like diy bio like you know you can make kombucha at home and i'm sure that an idea somebody might have is like well let's like engineer the organisms that go into making kombucha to do some property that you know maybe yeah We'll make it glow in the dark. There are things that you can do that are cool. There are luxuries, though, ultimately. And therapies are not luxuries. Therapies are necessities. And so I think if you can identify a necessity that's not a therapeutic, all of a sudden it becomes a little more clear. So can we turn plastic into something a little bit more palatable? That's a necessity. I would say that's a necessity. We just can't do it yet.
Speaker 1
I think it's a bucket of stuff. Yeah. I'm not meaning to like pick on Drew Endy. Fascinating dude. I should have him on the show. I only bring this up. I remember him being like this, having this huge celebrity moment for multiple years. And then it's like, I don't feel like I've heard anything in a long time, but maybe I'm just not paying attention to the right, right place. And I am not trying to, call him out on, it's, it's more reflective of this broader picture.
Speaker 3
Yeah, no doubt. Yeah. I, you know, I, it doesn't, to me, like, I, I think it's, it's definitely reflective of the state of biotech. I was like, there are cool things happening. So like I worked on gene drives for a long time. Yeah. Yeah.
Speaker 1
I was reading about all this. I mean, there's so many things to get into. I, I feel like I could keep you here forever. Um, I mean, so you were working on these mosquitoes and it was stuff people might have read about where you, you try to kind of sterilize the male mosquitoes and, and then, and then, yeah, try to, eradicate mosquitoes because they suck. And, and obviously I think this was a George in the church lab and then Bill Gates has had a mosquito thing for a long time. Although I think they were like zapping them with lasers. Um, but yeah, I feel like this was an idea that I kept seeing in the press was we're going to sterilize all these male mosquitoes. And then I know you've written about this, that it didn't, you guys made progress, but it didn't quite pan out the way you thought.
Speaker 3
Yeah. Uh, so, you know, just to, uh, to talk a little bit more about the principle there. So the idea behind a gene drive, so we call it a gene drive. What this actually is describing is like, can we drive a certain type of genotype into a population, a wild population that is super Mendelian? So like Mendelian inheritance, if we go back to like, middle school or whatever, high school biology. We all remember the monk. Yeah, the monk. And we do the Punnett squares and like, yeah, we do the crosses and the big A and the little, right. So that's Mendelian inheritance. So, you know, there are certain percentages we expect to be like the outcomes of that.
Super Mendelian inheritance is like if we take two parents and we cross them, it's going to be 100% every time. And what has allowed that is like CRISPR-Cas9. People have thought about this idea for a long time with other types of editors that were a lot harder to work with. But CRISPR-Cas9, all of a sudden the idea is like, well, can we use this to edit the embryo to make a generation generation? of mosquitoes. So like one idea that people thought of is, can we make mosquitoes infertile? Another idea that, so the one that I worked on specifically, the idea was like, can we make a, what's called evolutionarily stable gene drive?
So the idea is, can we introduce a gene into a mosquito that makes it like incompatible with carrying malaria? Such that like the ecological questions that people have, so like population crash, infertile, the idea is like you make a lot of these mosquitoes by injecting the embryos, make a ton of them that are infertile. And because mosquitoes mate one time for life, if you can make the males infertile, you can kind of basically like cut off the supply of new mosquitoes and you suppress the population. So how do we reduce Zika or malaria or whatever? Uh, well, one idea is we just, let's decrease the number of mosquitoes that exist in the world.
Previously, you know, in the U S we got rid of our mosquitoes by spraying chemicals everywhere. And we kind of got rid of malaria in this way, but we learned that's bad. So we are trying to be a little bit more known. Yeah. Just running through like the streets, like, yeah. So, um, you know, so that was the idea there is like, can we more ecologically, uh, in a more sound way, introduce like a species that can still, you know, we're gonna have mosquitoes. So maybe there's some organism that we don't know about that's really reliant on mosquito populations. Turns out like- Fuck them, man. Yeah, I know.
I'm not here. I don't work for big mosquito or anything like that. But yeah, so like what I worked on is trying to use CRISPR-Cas9 to introduce a stable gene drive into populations such that I can propagate something through nature and then you release them. However, however, it's really hard to do this. Biology is like, it doesn't work with you as well as you want it to. So as you know, I alluded to it earlier, very pointedly, maybe that like computer science is something where first principles help you a lot. Biology is not really one of these fields. Like you can first, you can think, you can reason through stuff.
You can go, well, if I use a promoter that expresses in the embryo, And I use a CRISPR-Cas9 system and I use a guide. It'll edit the thing and it'll work and it's going to be great. Yeah, it turns out there are these weird things that mosquitoes do that we don't understand. So, for instance, mosquitoes will like the way that they do a mating dance might work against you. Mosquitoes don't. It's not like you don't have mosquitoes you can introduce, for instance, into like Vietnam and have it move through Laos. The mosquitoes introduced in Vietnam might only stay in Vietnam. And it might be more specific than that.
The mosquitoes introduced in Hanoi might not ever move elsewhere in Vietnam. So it's like the scale problem is like one issue. Mosquitoes don't like mix homogeneously super well. Turns out mosquitoes can also kind of like have this sense of like, there's something not right about that mosquito. It's like, maybe it's like a hair too small. Maybe it's like cuticle abdomen is like a little too thin. Or the other. It's the other. Yeah, yeah. It's like, it's another, right? It's like, they can identify. It's like, they have this like uncanny valley. Like, that's not right. That's wild. Yeah. So it works against you.
It's not to say it can't be done. So, you know. I am very hopeful that somebody will solve these problems. And I worked with a lot of great people at Harvard who now started their own labs, working on different parts of the concepts there. But ultimately, there are a lot of things we don't know about mosquitoes. And that's hurting us towards getting that to work. And I think the Gates Foundation is trying to really think about the malaria problem in different ways. So now there's this really cool idea from a postdoc I worked with that at Harvard. Now he has his own lab at University of Georgia, Doug Patton, where it turns out you can just treat a mosquito net with an anti-malarial drug and you can treat the mosquito of malaria and it works.
And so, you know, that's cheap. It's very cheap. Is that better than editing mosquitoes? Like, I don't know, but it's certainly an idea. Other ideas people come up with is like there are these bacteria that exist in some species of mosquito that make it incompatible with carrying certain viruses. So Wolbachia is the bacteria. Can we take that bacteria from one mosquito? Can we move it to a different species and now make the species the super carrier of Zika, the 80s mosquito? Can we make it no longer able to carry Zika anymore? Yeah, I don't know. People are working on it though. So I'll just say like biology is just as tough.
It's a, it's like a tough, it's a tough game for tough guys. You know, it's not for the well-adjusted. You talk like someone who would be a biotech startup founder in that you're so
Speaker 1
knowledgeable about all this stuff and excited by these ideas. You are getting a medical degree, which sounds like a lot of time and effort. Why?
Speaker 3
Yeah. Why? Yeah. It's a good question. Um, You know, I was in George's lab. I fell in love with the idea of like genetic agents. So I worked on mosquitoes, but the idea that you can fix problems at the DNA level. And then I worked at the startup dyno and our idea was like, so we, I worked a little bit on editors, Georgia's lab. I learned a ton there. It's nice to be like on the left tail of the IQ curve and like a new environment. Like I was like the dumbest guy there. So you just ask like, wait, what is that? What does that mean? So I learned a lot there. Very intimidating place there. it was no everyone there is wonderful out of marble stone well you know folks like that the church lab alumni network nicest people on the planet yeah also a podcast
Speaker 1
guest there you go you're
Speaker 3
Just naming them, yeah. I will plug you. If you haven't heard that one, you should go listen to it. You should also listen to the Priscilla Chin. Fantastic interview. But so, you know, I worked on editors, and then there was a startup that was spinning out of George's lab, Dino, started by a graduate student of his, Sam Sanayi, and Eric Kelsick, the CEO. So they started a company, and the idea was like, well, let's work on this. There's a new thing people are doing in biology, which is like applying machine learning. to proteins. What is a protein that matters a lot? Well, viral capsids, viral proteins.
And can we apply those? Can we evolve the virus towards a thing we want it to do? Can we evolve the virus that goes to the liver to instead go to the deep nuclei of the brain so you can target certain diseases? And so I worked on editors, I worked on editing organisms, and I worked on delivery vehicles. And then at some point there, I was like, I think we're pretty damn close to deploying this in people. It was like a moment for me where I felt like I went to ASGCT, which is the American Society for Cell and Gene Therapy or Gene Cell Therapy, whatever. You get the acronym. So I went to a conference and you see all these crazy things people are doing.
It's like, wait, we're actually kind of solving genetic diseases for real. and i thought about like what you know what do i want to do and it was very compelling to me like i i would like to work at the intersection sort of the classical maybe like the way that uh maybe like doctors used to like work on still like don thomas uh who uh did the stem cell therapy like he was a he monk you could say i guess they didn't really have fellowships back then but he was a doctor and he also like worked on kind of like basic science, but like applied to people, translational sciences, as it's called. And that's something, yeah, I was like, I would like to do that.
But for gene therapy, it's like my thing that, like, I love this technology. And it also allows me to like touch on a lot of stuff. Like I can still use the protein biology, protein engineering, things that I learned at Dyno and also at Caltech. I can, there's a lot of stuff I can bring in. And it's like, it keeps my life interesting. And you're in, you're in Burlington, Vermont, the home of the greatest band of all time fish.
Speaker 1
Yeah. Yeah.
Speaker 3
Yeah. Burlington, Vermont. It's like the San Francisco of Vermont. Yeah.
Speaker 1
Um, all right. I'm going to let you go one second. You did just, I think we'll close on this, this tweet, this, whatever we call it now, um, which is whatnot. So it was about vaping. So like as a doctor, okay. You're, you're, your post, your tweet, if I remember it was that you're seeing these youngsters come in who are like inhaling, you know, just inhumane quantities of nicotine every day. This was something you saw anecdotally as a practicing doctor.
Speaker 3
Yeah. Uh, you know, it, It was like, it broke my brain a little bit, to be honest with you. I, you know, cigarettes, cigarettes are bad. Kids don't smoke. Now people are talking about like Zins, ultimately, like, you know, is nicotine at certain quantities like bad for you? Like, yeah, you know, probably... Jury's still out, but like, it's probably not that bad. Maybe it's like good for short term stimulant. Right. Right.
Speaker 1
Like blood pressure. Yeah. You know, there are things that are probably bad and we'll learn them later.
Speaker 3
Maybe they're good. Who knows? Um, but it's like at a certain quantity. It's like a pack of cigarettes. Um, we'll say it's like 20 cigarettes in a pack, one to two milligrams of nicotine per cigarette. So we're sort of at like, depending on the brand, 20 to 40 milligrams of nicotine in a pack of cigarettes. Now, what I've observed kind of is, it's really even hard for me to like mentally process, is like a lot of young kids are like going through these like, nuclear grade vapes, like made in China, like the disposable one.
Speaker 1
Yeah. Those big cartridges. Yeah.
Speaker 3
Yeah. Big hand size. Yeah. Geek bars, elf bars. Like they have these like really, you know, bright colors and you know, they got like Tamagotchis on them or whatever. Like it has like more compute than like the Apollo, you know, like it's, these things are like insane. It's like, is there like an H 100 in this? Like, and they're going through them two, three of the, you know, two, three a week. And every single one of them is like a thousand milligrams of nicotine. It's like smoking 20 packs a day. Our concept of, you know, what is safe is out the window. Like, is coffee bad for you? Like, no, it has some mild effects. But if someone is drinking 20 cups a day. A bathtub. Yeah. You're like, you need to, you know, relax, big dog.
Speaker 1
And you. they told you they're doing two or three a week or you're seeing they're like they're like in the hospital room like yeah yeah i've observed like patients will they get like very antsy like nicotine withdrawal it's like it's not a pretty thing it's not like you don't have a seizure but like people are irritable and like they're pacing around and like we give them nicotine pads so if you can't smoke in
Speaker 3
the hospital yeah um for a lot of good reasons But we try to help the people who do smoke. Like, you know, if you smoke cigarettes and you're there for like a blood infection, we're not there to deal with the cigarette smoking today. We're there to deal with the blood infection. So smoke cigarettes, smoke a pack a day, we give you a patch. It's roughly equivalent to the same amount. Um, how do you get to a doc? You don't have enough patches to get me where I need to be. Yeah. And it's like, they just like, they want to go. Like they have, in this particular case, it was like a patient who he had a fairly serious medical problem to address. It was like, yeah, I just like, I just need to go like vape. And like, how do you sit there and go like, No, you do not. But also like, I guess we could give you, I can give you nicotine gum, give you patches. I can give you a lot of juice. I can give you a lot of things, but like this is going to add up to like a 10th of your usual intake.
Speaker 1
That was the part of your post, the end part, if I'm quoting it right from memory was something like, I don't even know how you begin to quit. I don't, I really don't.
Speaker 3
I really don't. It's like, yeah, I have no idea. It's like, if someone's drinking like two liters of vodka a day, like you're going to go, well, it's going to be really terrible for a good amount of time. Yeah. You need to lock in for that. It's not going to be good. And I'll just sort of, I worked at Caltech. Um, Henry Lester was, uh, uh, he's, he's a Caltech professor in the bio department there. He's been at Caltech for a long time and he made his career, um, doing like nicotines, you know, nicotine, understanding like what are the effects of nicotine in a cell, cell stress, how does it work?
He's a neuroscientist and he's been doing this for a long, you know, 40, 50 years. Um, And I remember like last summer I was having lunch with Henry and he said something to me like I didn't I was like, yeah, I don't agree. He was like, I think vaping is is bad in a worse. Like I think vaping could be worse than cigarettes. And I was like, I don't know. I think cancer like lung cancer is pretty bad. Turns out smoking cigarettes causes other types of cancers, pancreatic cancer, esophageal cancer, mouth cancer. It causes a lot of types of cancer at decently high rates if you do it enough as people were for a long time.
I don't agree. He might be right. I, I, yeah, he might be right.
Speaker 1
Right. Cause you're saying, you're saying when you smoke a cigarette, you have all these carcinogens doing all kinds of stuff. The idea with vaping was, was that maybe it's cleaner and you're, but you're talking about this just like quantum, maybe there's just this addiction, this, this all consuming. It's, It's tough because I can't point to a thing and be like, I think this is why it's bad. I could say I don't think that degree of vasoconstriction is long-term good for your arteries.
Speaker 3
My guess is it's not good, but I don't know. I don't have empiric evidence to point to. But Like I could, you know, I could probably tell you if you, if you did this for 10 years, you're going to not have, your bad things are going to probably happen to you. You know, they cracked down on Juul and Juul was like not one of those big reservoir things.
Speaker 1
And it was like, it was like a, I think it was still a lot easier to just puff on your Juul all day than a cigarette, but, but it was, it was like, it was kind of measured and, and like, It actually felt like a company. I'm slightly talking out of my ass because I don't know, but it felt like a company that was under all this pressure to do this well. Yeah. So like, hopefully, you know, the stuff you're talking about, you're just like buying it at some head shop and who knows who makes this thing and what's in it.
Speaker 3
And they're, yeah, they're, I mean, they're the difference like Juul. Juul had problems. Like they, you know, they were trying to do the flavor. They recognized the flavors were things that kids were into. And so it sort of defeated the purpose of the off-ramp from cigarettes into vaping. And now it became like the on-ramp to cigarettes became like using a mango Juul pod in like your high school bathroom. For sure. And so, you know, that was bad. But the amount of nicotine in those devices was ultimately like a little bit more comparative to what you could buy in a pack of cigarettes.
Speaker 1
Yeah. I just think it's, there's some irony that we cracked down so hard on them and but like you were still able to sell all the others we're like we'll take the one company that make an example out of them because they did the flavor pods and but you know I mean they were bought by Philip Morris I think like they were under pressure to kind of do it right Philip Morris isn't going to sit there and just take on like even more litigation and things like that And then now, you know, well, I think you can buy it in California again, but it was banned for a while, and you could still buy the other. And they don't have flavor.
Speaker 3
Like, that's the thing. Like, the resolution to that was like, okay, you still buy Jules. but you can't buy flavors and they might sell like tobacco flavor, but it's like nothing flavor or mint or whatever. It's like nothing really. Like kids don't love, the kids don't love tobacco flavor, but they love cool raspberry crush creme brulee. They love that stuff. And, uh, And we're also learning like a lot of these vapes unregulated. We have no idea how they're made. And they have like heavy metals in them. Like I just read a paper or a study a few months ago where they showed like all these heavy metals in these vapes. And it's like, I don't think like... yeah i think there will be some reckoning uh there um so get working on those gene therapy buddy uh thank you so much for coming on what should we point people to that you want them to to check out uh you know you can you can read my you can read my blog if you want bio stuff or you know look up on Substack yeah onSubstack um yeah i don't know i just i yeah i like talking to people but you're a good tweeter I'm a good poster poster.
Speaker 1
Yeah. And then, and then please definitely go to core memory and read the story. Um, terribly. I can't remember the title off hand. It was the gene therapy playbook. Yeah. Yeah. Um, And, uh, and I know, and thanks to Brendan Burrell, who was the editor on that story. Um, and yeah, man, thanks so much for coming in. Yeah, absolutely. Yeah. This was a great time. Thank you so much. Where are you going to go? You're going to go run off with a lot of AI people now or what?
Speaker 3
Yeah, I am. I'm going to go do this. I'm going to like something at situational awareness later today. Um, and from, from there, what happens? I'm not sure. Okay. Yeah. Yeah.
Speaker 1
All right. Well, thank you for checking out from Vermont. We really appreciate it. And we hope to have you write for us more. Yeah, absolutely. Thank you so much. We got a commitment. All right. All right. Thanks everyone. The Core Memory Podcast is hosted by me, Ashley Vance. It is produced by David Nicholson and me. Our theme song is by James Mercer and John Sortland. And the show is edited by the John Sortland. Thanks as always to Brex. Anyone Ventures for making this possible, please visit our Substack, YouTube, and podcast channels to get more of what Core Memory makes. Thanks, y'all.