Reversible cryopreservation is achievable because we already do it for small tissue — the remaining question is scaling up.

(00:00:00) What if you could take someone who is on their deathbed and find some way to hibernate them until the sort of critical cure for the disease comes online? (00:00:08) The ability to freeze time for humans. (00:00:11) I didn't actually think that was something you could go work on. (00:00:13) So apparently it is. (00:00:15) Our long-term goal is reversible full-body cryopreservation for medical hibernation. (00:00:18) But in the near term, what we work on is reversibly cryopreserving single human organs to help transplant patients get organs more efficiently. (00:00:24) Making time not a variable changes the whole paradigm. (00:00:26) One thing I love about the field of cryopreservation is I think like the (00:00:30) The problem speaks for itself. (00:00:31) Water expands when it forms ice. (00:00:34) That's just hard for your tissue to take without substantial damage. (00:00:37) And the cool thing is that there's sort of a temperature below which ice formation stops happening. (00:00:41) So basically, if you can traverse and you can get below that without ice formation, then you're good. (00:00:46) We already reversibly cryopreserved tissue, including human tissue, all the time. (00:00:50) And we do it for very long time periods. (00:00:52) There are kids who were literally cryopreserved for 30 years as tiny embryos. (00:00:56) And so the main question is not, is this possible to do at all, it's is it possible to scale up? (00:01:00) Even if that's true, why don't you think it's been worked on? (00:01:09) Hi listeners, welcome back to No Priors. (00:01:11) Today, I'm really excited to be here with Laura Deming, previously the founder of the Longevity Fund and now the co-founder and CEO of Until. (00:01:19) We're going to talk about how Until is progressing the frontier of reversible cryopreservation, or freezing living things and waking them back up, beginning with human organs progressing to small animals and hopefully making progress on the whole body. (00:01:35) It sounds like science fiction, but we'll talk about some of the scientific challenges, where we are today, and the implications if this is possible. (00:01:42) Thanks so much. (00:01:43) Welcome, Laura. (00:01:44) Laura, thanks so much for doing this. (00:01:46) Yeah, thanks for having me. (00:01:47) I've been so looking forward to this since our Pantheon Watch sessions. (00:01:51) We're talking about upload and the nature of consciousness. (00:01:53) But one thing that you don't know is that my like very long ago wished for technologies that I wanted to exist were telepathy. (00:02:02) upload and the ability to freeze time for humans. (00:02:05) I didn't actually think that was something you could go work on. (00:02:08) So apparently it is. (00:02:10) How do you end up working on that or being interested in longevity at all? (00:02:14) There's 2 different questions. (00:02:16) So yeah, I come from a longevity background, but in my mind, like reversible cryopreservation is applicable a bit outside of that as well. (00:02:22) I don't know. (00:02:23) I mean, I think I'm really obsessed with areas that feel like they should be worked on but aren't. (00:02:30) And (00:02:31) when I was a kid, I think naively just growing up, that seemed really obvious for longevity. (00:02:36) And it was really surprising to realize that, it's not the case that most people like are working on that explicitly as a goal. (00:02:42) And in fact, that like, it's kind of like, I think longevity and aging kind of occupy this weird realm where because they're not. (00:02:49) explicitly diseases in a way that's fully socially recognized yet. (00:02:52) They're not seen as valid to work on, but like that's not really for, I think, technical reasons on some level. (00:02:57) It's more for like classification reasons because like, you can extend the lifespan of like sort of many different organisms using technology. (00:03:06) And how much can you do that in humans? (00:03:07) We have no idea. (00:03:08) And it could be very small for technology, but sort of like I think longevity is interesting because it feels like an area where there's a social blind spot around something. (00:03:16) And I find those very interesting. (00:03:17) Perhaps I was just, I'm sure this is true, not very observant as a school-aged child, but I don't think I even understood aging was like a concept that I should consider at all. (00:03:29) And so how did you end up thinking about it in any depth? (00:03:32) I grew up in a pretty... (00:03:34) odd setup. (00:03:35) So, I was in New Zealand, I was homeschooled. (00:03:38) I didn't really have a, like I didn't go to a normal biology class. (00:03:42) I was kind of, by myself in the house. (00:03:44) I imagine you like staring at a field of sheep and then being like, someday we're going to get old. (00:03:48) I should do something about this. (00:03:50) that would have been the farm that we had for a little bit. (00:03:52) But I remember one thing that really stood out to me was at some point when I was a kid, I was thinking about how long people in my life were going to live and how old they were. (00:04:00) And for some reason, it made a lot of sense to me that everyone should live until they were 10 years old and then die immediately at 10 years old. (00:04:05) Like I just, like that seemed like some hypothesis. (00:04:07) I didn't really know how old people were. (00:04:09) And so working backwards, I was like, oh, my dad must be like, you know, maybe 8 and my mom's everything, so like maybe 7. (00:04:14) But I think one thing that was really striking was realizing that we don't all live until a certain age and then we die. (00:04:19) In fact, we don't know like what determines how long we live. (00:04:24) Like that was very interesting, right? (00:04:25) This idea that like, almost like if we all lived until 10 years old and then we died immediately at 10, like I would feel much more confident with the idea that like longevity is some kind of immalleable like hard limit. (00:04:33) But they do that like there was uncertainty about that. (00:04:36) It was really interesting. (00:04:36) And it's sort of like, what are the factors behind that uncertainty? (00:04:39) So I'm going to fast forward through a bunch of like lab work you did (00:04:44) and going to MIT and being a Thiel fellow, like why a longevity fund? (00:04:48) Me just being very literal. (00:04:49) Like at the time that I was interested in longevity, I think if you ask the average person in the field, like what's the big problem, most people would say, well, we just can't get enough funding for our projects. (00:04:59) And so that's the big problem. (00:05:00) And I just took that literally like when I was a teenager. (00:05:02) So I also have that problem. (00:05:04) Yeah, I just like literally like I should just get a lot of money to like help push longevity drugs forward. (00:05:09) And the name for that happened in venture capital fund, but it definitely wasn't working downstream of the idea of venture capital. (00:05:13) Like that came (00:05:14) after this idea of just getting money for projects that should have money and then. (00:05:19) You did this for a number of years. (00:05:21) What triggered the sort of change in how you were going to spend your attention to cryopreservation? (00:05:27) I think just like cryopreservation is one of the coolest, like I think it's not like to pardon the pun, but I think it's one of the coolest, most interesting, like best problems ever. (00:05:34) Like I think like, I mean, if (00:05:36) from so many different angles. (00:05:37) Like I think you've got to have a counterfactual impact. (00:05:39) Like if you're obsessed with just technical delight, like just Rosh, your technical interest and diversity and the logical parts of the problem. (00:05:47) And then also I think from perspective of social impact, like I'm just, it was just, it's such an interesting problem from like how it's perceived and then like what are the different factors of that. (00:05:55) It was like 0 to one. (00:05:56) It was like the, I remember just seeing the problem clearly for the first time. (00:06:00) I mean, a lot of the people in my life, I think had (00:06:03) I've been aware of it, but I just, it took me so long to really, I think, see it clearly for myself. (00:06:08) But then it was just like 0 to one of like, this is the only thing that I could imagine pouring the next decade of like my work into it. (00:06:14) That was after I kind of done kind of the first set of work with fun. (00:06:17) I was kind of like thinking like, what is the next like big thing? (00:06:20) With your co-founder Hunter, was it like an immediately obvious, yes, we should work on this together? (00:06:26) He's one of the few people where if I ask him a question, like he'll come back the next day (00:06:31) and give me an answer. (00:06:32) He's literally thought through from first principles, like what the correct answer is. (00:06:36) Like, I remember at some point I realized that he'd written a doc on like the principles of like, you know, cry preservation. (00:06:41) Like most people, you know, like, would write that from the literature or kind of like citing different sort of sources at various levels of granularity. (00:06:48) Hunter went back and like re-derived fundamental laws of stat mech as part of like, this is what you should know in cry preservation doc. (00:06:55) Like, I really admire how much he builds up from (00:06:59) really simple models to try to create coherent, like technical pictures that are more complex. (00:07:04) He's like the most fun, interesting, best person to work with ever, for sure. (00:07:08) And it wasn't a hard sell, like we should go work on this problem in particular. (00:07:11) It was not a hard sell, but it was not in an interesting way. (00:07:13) So I think one thing that I like love about the field of crowd preservation is I think it's very compelling. (00:07:20) Like if you like hard, like it's one of those things where if you (00:07:24) the problem speaks for itself. (00:07:26) And so I remember telling country about it in our first call and he basically was like, I don't buy it. (00:07:32) He told me later that he didn't tell me in the call. (00:07:35) But then he went off and thought about it for a couple of days and really thought about what we knew about ice formation, like did some basic back of the envelope math and came back and he was like, oh, wait, this seems like actually, or like this seems like it's in the realm of possibility in a way that is very different from my initial intuition. (00:07:51) And that sort of like conjugation is just so interesting. (00:07:54) Interesting to me. (00:07:55) Maybe it's useful to zoom out and just say, like, what is the goal of Until? (00:07:59) I would think about our goal as trying to create a new form of critical care. (00:08:03) The example that I would give that is sort of the core of the company is there's some years where certain diseases such as like metastatic melanoma go from being, like, sort of in a single year, like metastatic melanoma went from being something that you had like a six to nine month prognosis, like less than a year of expected survival. (00:08:19) to with new combination immunotherapies, you might have a decade plus of expected survival, or 50% of people sort of surviving over a decade, without getting sort of death from melanoma. (00:08:31) In fact, they're starting to have other things at that point. (00:08:34) The tagline is like, single years can make the difference between a patient dying of terminal illness and like living long enough to make the critical cure. (00:08:40) But right now, there's no way to press pause on their biological time. (00:08:43) It's like, what if you had an ambulance to the future, right? (00:08:45) Like, what if you could take someone who is on their deathbed and find some way to just sort of hibernate them basically until... (00:08:54) the sort of critical cure for the disease comes online. (00:08:56) And in this context, we're not talking about necessarily decades or kind of like much longer than that. (00:09:01) Initially, it's just kind of in the context of like when there's a window where you could imagine like a critical trials being done for a drug that were they eligible for it could make a huge difference for their disease. (00:09:12) To give an example of like the need to, it's like my sort of co-founder's father-in-law (00:09:18) had this happened to him in the sense that he got a sort of advanced cancer that would have been treated or treatable or addressable by a therapy that came out basically a couple months after he was no longer eligible for the therapy. (00:09:31) And he like missed, the critical clinical trial by like, a couple of months. (00:09:34) And so it's sort of like that level of urgency that like someone in that position shouldn't have to, miss a critical therapy because there's a couple month difference in like when they got their sort of disease and when the therapy came available. (00:09:48) And from a just like product perspective, that means he needs to do whole body cryopreservation. (00:09:54) Yeah, so to give context on technically how we can go about this problem, so our long-term goal is reversible whole body cryopreservation for medical sort of hibernation. (00:10:03) But in the near term, what we work on is reversibly cryopreserving single human organs to help transplant patients get organs more efficiently. (00:10:10) I want to come back to all of the technical challenges here and where we are and what you think the next milestones are. (00:10:15) But because you describe it in the context of medical use, like I'm going to be honest, as a kid, I was like, well, I want to be able to freeze time because I want to be able to go to Mars too. (00:10:26) Or you said, because you've worked on longevity with this perspective, like aging is, it could be considered a disease, a health state we should work on that is credible medical science. (00:10:37) So how do you think about those other use cases? (00:10:39) So an interesting thing when you start to think about like actually applying this technology is sort of what's the experience of the person, not just kind of technically for the disease, but socially. (00:10:49) So like one of the number one reasons most people wouldn't do medical hibernation, and especially that it wouldn't be like for the most part of recreational thing is that (00:10:56) I think a lot of people view themselves in part defined by their social contacts. (00:11:00) So when you say recreational thing, you mean, because you just, you think of that as going to Mars as a recreational use case. (00:11:08) Oh, no, sorry. (00:11:09) I think this case, what I'm thinking of is, some people might imagine that they would love to skip into the future just to see what happens, and maybe, have their same amount of number of years of life, but like, (00:11:19) have the future shifted by some amount of time. (00:11:21) But the number one reason that most people wouldn't do that and are also just very wary of the idea of hibernation for themselves is, I can't take everybody with me. (00:11:30) Yeah, exactly. (00:11:30) It's a city that like you kind of find yourself by the people around you. (00:11:34) And so I think like, those kinds of use cases, like going to Mars, those are all kind of things that could happen and could happen with or without or could happen (00:11:42) will require technology for certain definitions of mind. (00:11:45) But I think the thing that a lot of people will face is just like this question of like, is it worth traveling so far to give up my current social context? (00:11:51) And that will put a limit on like how much people want to use this for like, I think there's like a real cost that kind of you incur. (00:11:57) And so it's, in my mind, it only makes sense for like really serious use cases initially. (00:12:01) where you literally would die, or because you're kind of putting on the line like your current, all of your current social reality and how that will evolve without you versus like this other thing that you might want. (00:12:10) But I mean, a lot of people might want to go to Mars in that context, which I think I'm a little, yeah, the cost is pretty significant. (00:12:15) It's very hard to know what we want though. (00:12:16) Like a lot of people might make that decision whether or not their ultimate happiness is higher. (00:12:23) Maybe you can just break down how you think about the challenges scientifically, right? (00:12:27) Versus I think to maybe even Hunter's original reaction, like, sounds like science fiction. (00:12:33) I don't know if you can go work on that thing. (00:12:35) Yeah. (00:12:35) Right. (00:12:35) And so if it's, you know, crystal formation or whatever the set of challenges and what sequence you think you should solve them. (00:12:43) I see. (00:12:44) Yeah. (00:12:44) Maybe I can just give a series of facts that I think. (00:12:50) together sort of make the problem super interesting. (00:12:53) So one fact is that ice formation is a stochastic process. (00:12:56) So if ice just formed unilaterally in any given material past a certain point of temperature, like, you know, just like go from zero to 1, it's like 100% ice, like that might be kind of hard to think carefully about technically, but ice sort of forms through a process of (00:13:14) random nucleation and then extension. (00:13:16) And this is cool because you can modulate the sort of nucleation, the rate of nucleation and extension to then modulate the probability of ice formation. (00:13:26) And because it's probabilistic, if you can do that well enough and you can sort of spend minimal time in the temperature range where (00:13:32) ice can nucleate, then that gives you a shot at sort of preventing a lot of ice formation. (00:13:37) So like the number one tagline would be like, avoid ice at all, or like sort of avoid as much as possible. (00:13:42) There may be some, like, I think some people might be working on technologies to cryopreserve with some sort of like ice formation, but we're focused on regimes where you're basically trying to avoid as much ice formation as possible. (00:13:51) Just for the non-biologists, you know, ice formation is bad because it breaks all the cell membranes. (00:13:56) Yeah, so ice formation is bad because ice expands, water expands when it forms ice. (00:14:01) And that's just hard for your tissue to take without substantial damage. (00:14:05) So you want to avoid ice formation. (00:14:07) And the cool thing is that there's sort of a temperature below which ice formation stops happening. (00:14:13) So basically, if you can traverse, let's say, you know, going below 0 degrees or less, through to around minus 130, and you can get below that without ice formation, then you're good. (00:14:24) And (00:14:25) The interesting is at that point, you're good for quite a long time. (00:14:28) So there have been human embryos that were reversibly cryopreserved for, the latest record was over 30 years, and then rewarmed and sort of viably used to create pregnancy. (00:14:40) And then, you know, sort of like there are kids who were literally cryopreserved for 30 years as tiny embryos. (00:14:46) And so that's the last thing which was very surprising to me, which is that like we already reversibly cryopreserve tissue, including human tissue, including whole body human tissue at that very, very small, like you know, in a couple hundred cells stage. (00:14:58) all the time. (00:14:59) And we do it for very long time periods, which I honestly like what a first principle's been stuck on. (00:15:04) Is this possible to do at all? (00:15:05) Is it possible to scale up to a large complex biological system that has a lot of vasculature where you're dealing with different material properties, where you have to think a lot about like perfusion and how to sort of diffuse chemicals in and out and how to (00:15:18) get heated out quickly enough. (00:15:19) So the idea that you could pause all molecular motion and then randomly restart it, and even a cell would survive that, from which principles to me used to seem crazy. (00:15:26) But we know that works. (00:15:27) Yeah, it's like, we just, we already, like, scientists just tried it and it worked. (00:15:31) And so now the problem is like scaling that up and doing it in a way that's compatible with like, you know, tissue health. (00:15:36) Yes, I think one of the things that was most wildly surprising to me, like being in your lab a little while ago, is how much it looks like people were working on what I'd consider to be like engineering problems. (00:15:48) It's like, how do we get something to warm quickly and safely enough versus let me go work on this therapeutic. (00:15:56) Yeah. (00:15:56) So actually, I feel like there's just part of the problem that I've been trying to explain externally for a long time. (00:16:02) Every time I try to explain it, I think it comes off as like not specific or something, but it is actually one of the core reasons why I think the problem is interesting to work on, which is that like you can trade off like engineering difficulty and biological difficulty to a non-zero degree. (00:16:16) Like not 100%. (00:16:16) You can't just use engineering (00:16:18) to solve the problem. (00:16:19) You absolutely have biological questions, and those questions could come out in the negative for some of these cases. (00:16:23) So that's not saying you can just make an engineering problem. (00:16:26) You can make your life easier on the biology front by building better engineering tooling. (00:16:32) And the fact that that's possible is a huge deal. (00:16:34) That is not true for most problems in biology. (00:16:38) And it gives you a lot of leverage on the problem. (00:16:40) This is probably interesting to maybe only like 5% people watching this, but I think that I'm obsessed with is just the idea like temperature is such a beautiful conceptual tool, right? (00:16:46) It's like temperature as an idea, something that in physics took physicists hundreds of years to come up with. (00:16:52) It links like molecular motion to a high, like a single high level measurable parameter. (00:16:58) And just tuning temperature like sort of tells you about almost like the relative passage of time of like molecules at the nanoscale. (00:17:04) Like that's a highly non-trivial sort of (00:17:07) conceptual lever to have on a problem. (00:17:09) And in biology, one of the biggest problems is like, it's really hard to find powerful conceptual levers on sort of like for like nanoscale, for manipulating like the nanoscale, that have anything approaching that degree of sort of leverage. (00:17:23) Basically, what that gives you is like, you can apply a lot of theoretical sort of, a lot of theoretical toolkit used in physics to model (00:17:33) parts of this question in ways that are actually useful. (00:17:36) And it is just not true that you can use like equations from physics to think usefully about almost any other problem in biology. (00:17:43) There are projections like in the context of like medical devices, but like in a context where you're talking about like changing the course of terminal illness, which this one is interesting because this doesn't unilateral change it, but like it gives you the possibility of some more time. (00:17:58) I think it's one of the most important things to understand about the problem. (00:18:01) I don't yet understand how to explain it in a way that is clear, but I think it's like one of the most important things to understand about the problem. (00:18:05) Yeah, maybe if I think about actually applying it like just very concretely to what you are doing, like if it is... (00:18:13) challenging from a organ preservation biology perspective to have a organ reheated or sorry, rewarmed like evenly throughout, then maybe the thing to do is to like change the surface area to volume ratio of like your heating device or distribute the heat in different ways without like changing your understanding of the biology, but just with new devices and technologies that you invent from the engineering perspective. (00:18:40) Like that actually seems like a very simple example. (00:18:42) I realize you're implying a (00:18:43) like more fundamental view of like why temperature is just such a interesting framework to be working on from both an engineering perspective and a biology perspective. (00:18:53) And there's like trade-offs where you put your effort here. (00:18:56) But I think that's actually, that's something that like didn't, did not occur to me at all coming into your lab and learning more about until I was like, oh, it's actually like a, to some degree, much simpler problem than I understood to the point of like, well, if you can just reduce (00:19:13) and preserve, reduce and increase temperature in these ways that are perfect through this organ is going to work, right? (00:19:21) Yeah, so let me restate that and then give one caveat just to make sure that, like, I can read correctly. (00:19:28) So a way that you can talk about the trade-off between engineering and biology is that, like, with engineering, you can modulate cooling and rewarming rate, just to some, to certain extent. (00:19:38) And then that can then change like how much what we call a cryoprotective agent or kind of chemical that modulates ice formation you add to the system. (00:19:46) And you want to minimize the concentration of that cryoprotective agent. (00:19:49) So that you can be toxic. (00:19:50) Exactly. (00:19:51) And so as you increase cooling and rewarming rate, AKA spending therefore minimal time in kind of the danger zone of ice formation, you can correspondingly decrease the concentration of cryoprotective that you're putting in. (00:20:01) But if you could instantaneously cool and rewarm, then you wouldn't have to put any CPA in. (00:20:05) But that's not something that we're doing. (00:20:08) default assuming is feasible for a large system. (00:20:11) So there's still always going to be a component of biology, AKA like how tissue responds, especially from a toxicity perspective to like a new chemical agent. (00:20:18) Where are you now in this progression? (00:20:21) Like should I think of it as like there's a kidney and then it seems like quite a large jump to a small animal, but maybe it's not? (00:20:30) so we work on the two in parallel. (00:20:31) So we both work on scaling up preservation and rewarming technologies to kind of human organ scale. (00:20:38) And also in parallel, we work on sort of whole rat reversible hibernation and in terms of technologies over from sort of what we learn on the kidney side into the right context, as well as doing things specifically for rat. (00:20:53) When you started the company, did you have a timeline in your own mind? (00:20:56) So I think initially I was like, (00:20:59) We could maybe make some progress and hopefully make some good products, but the idea of full body cryopreservation felt like that would be really far out if that was possible. (00:21:07) I mean, I think it, I definitely still wouldn't put like a near-term panel on it, but I feel like we have a much clearer roadmap, at least to get, to begin to get there. (00:21:16) And the first steps seem faster than I would have imagined. (00:21:20) if that makes sense. (00:21:21) But the big unknown to get to whole body reversible is the brain. (00:21:26) It's unclear, like the brain can withstand a lot of change and does withstand a lot of different types of damage or change with age, for example. (00:21:36) But it's unclear whether, like what kind of injury the brain could sustain in the context of like a whole body reversible preservation protocol and then like what level of fidelity it's possible to do. (00:21:45) So like that, to be clear, is a big unknown on the neuroscience side. (00:21:48) How do you recruit (00:21:50) and lead in a company that has, let's say, an unclear timeline around a really big scientific goal like that, in terms of both finding people that are the right fit, motivating them, and how do you think about urgency in that context? (00:22:05) I mean, I think we have a pretty clear timeline for our first product, which is like, get reversibly cryopreserved. (00:22:10) So basically, transplant patients today surprisingly frequently miss organs that are en route to them because there's a timing problem. (00:22:17) So like, you know, organs expire very quickly after they become available from a donor. (00:22:23) That's very unfortunate. (00:22:25) Yeah. (00:22:26) And it's crazy because they're like one of the most precious resources we know of, and yet like people regularly charter private jets. (00:22:32) like you're putting a surgeon on a private jet to go pick up an organ, get it back to the patient in time. (00:22:36) And you're doing all that at the last minute and scheduling the patient for surgery at the last minute. (00:22:39) So the patient has to wait within like a two-hour radius of a transplant center with a pager on them or like a notification device at all times. (00:22:45) And so the first product that we're aiming towards is just, being able to pause time for the organs. (00:22:49) The patient doesn't, so you can take as long as you need to get the organs to the patient. (00:22:54) And that's very near-term, like that's not long-term ethical. (00:22:56) That's like we're aiming to get that into like pre-clinical studies and into the clinic as quickly as possible. (00:23:01) So I think it helps to have a very concrete goal that clearly is relevant to our long-term goal, right? (00:23:04) I think another thing that was inspired us about that product was like, if we're at all talking about whole-body reversible cryopreservation, (00:23:11) And we can't make a dent on that problem. (00:23:13) There's no version of it not going through it. (00:23:15) Yeah. (00:23:15) It's like, if you're serious, that should be doable. (00:23:18) And if you can't do that, then you're not the company to do the long-term thing. (00:23:21) So it was nice for us to have a very clear benchmark for ourselves of, are we correct that this is a tractable technology on that scale? (00:23:30) I guess, is it a challenge to lead beyond that, because everybody understands the company has a broader mission? (00:23:36) Or is it just focus on step one? (00:23:38) I think there's the possibility that's difficult, but I think right now I feel good about sort of our ability to compete around that, which is like, I think if we were like, It's 100% possible to do full-body risk of corporate preservation, there's no question we're certain, like then we would just be bullshitters and then we, like, we wouldn't be able to recruit because it's, you know, there's a lot of technical risk and there's a lot of uncertainty between here and there.