WEBVTT - Mass-Producing Stem Cells to Cure Disease 0:00:15.356 --> 0:00:15.796 Pushkin. 0:00:20.396 --> 0:00:25.476 There are these amazing cells in tiny human embryos. The 0:00:25.556 --> 0:00:29.876 cells are called pluripotent stem cells, and they're amazing because 0:00:29.996 --> 0:00:33.996 they can become any kind of human cell, a red 0:00:34.036 --> 0:00:37.596 blood cell, a skin cell, anything, any cell in your body. 0:00:38.396 --> 0:00:43.076 But pluripotent stem cells only exist for the first fourteen 0:00:43.196 --> 0:00:48.276 days of embryonic development. After that, they're gone forever. At 0:00:48.316 --> 0:00:51.396 least we used to think they were gone forever. And 0:00:51.396 --> 0:00:55.116 then about twenty years ago, researchers figured out how to 0:00:55.156 --> 0:00:58.876 take regular cells from adults, blood cells or skin cells 0:00:58.956 --> 0:01:01.956 or whatever. Take those cells, bring them into the lab, 0:01:02.036 --> 0:01:07.116 and then turn them back into pluripotent stem cells. These 0:01:07.156 --> 0:01:12.036 cells are called induced pluripotent stem cells, or ipsc's, and 0:01:12.236 --> 0:01:16.116 the possibilities they present for human health are both kind 0:01:16.116 --> 0:01:20.116 of obvious and awesome. If a person has a disease, 0:01:20.556 --> 0:01:24.556 you could use that person's own cells to grow any 0:01:24.676 --> 0:01:27.196 kind of new cells that they might need. You could 0:01:27.196 --> 0:01:30.516 grow new brain cells for Parkinson's disease, or new heart 0:01:30.596 --> 0:01:33.996 muscle cells for heart failure, or new bone marrow cells 0:01:33.996 --> 0:01:37.876 for leukemia patients. It has taken a long time to 0:01:38.076 --> 0:01:40.516 put that dream into practice, and in fact, it's not 0:01:40.596 --> 0:01:45.116 really solved yet, but researchers are getting close. A bunch 0:01:45.196 --> 0:01:49.036 of clinical trials are now underway using iPSCs to treat 0:01:49.076 --> 0:01:54.036 everything from Parkinson's disease to cancer to macular degeneration. But 0:01:54.276 --> 0:01:57.956 even if these clinical trials are successful, there will be 0:01:58.156 --> 0:02:03.396 another problem to solve. Turning a patient cells into iPSCs 0:02:03.516 --> 0:02:06.636 and then into whatever kind of cells they need takes 0:02:06.916 --> 0:02:10.876 months of work by highly trained science. It costs hundreds 0:02:10.916 --> 0:02:14.556 of thousands of dollars for each patient, and so even 0:02:14.556 --> 0:02:18.036 if those clinical trials are successful, the process of making 0:02:18.036 --> 0:02:20.756 the seals will still be too expensive and too labor 0:02:20.756 --> 0:02:24.676 intensive to ever benefit you know, millions of patients. So 0:02:25.036 --> 0:02:27.436 if the dream of IPSS is going to come true, 0:02:27.836 --> 0:02:30.956 somebody needs to figure out a faster, cheaper way to 0:02:31.076 --> 0:02:40.276 make them. I'm Jacob Goldstein and this is What's Your Problem, 0:02:40.436 --> 0:02:42.156 the show where I talk to people who are trying 0:02:42.236 --> 0:02:46.236 to make technological progress. My guest today is Nabiha si Client. 0:02:46.436 --> 0:02:49.596 She's the co founder and CEO of a company called Selino. 0:02:50.396 --> 0:02:54.596 Her problem is this, how can you make iPSC therapies 0:02:54.676 --> 0:02:59.756 quickly and cheaply? Before she got into the induced pluripotent 0:02:59.836 --> 0:03:03.396 stem cell business Nibiha was studying to be a physicist 0:03:03.636 --> 0:03:05.556 and she loved physics. Was going off to get a 0:03:05.556 --> 0:03:09.636 PhD in physics at Harvard, but just before she started school, 0:03:09.796 --> 0:03:12.436 her grandmother died, and she told me that had a 0:03:12.476 --> 0:03:15.476 really profound effect on how she thought about her research 0:03:15.636 --> 0:03:16.476 and her career. 0:03:17.516 --> 0:03:20.916 My grandma died due to severe diabetes that was not 0:03:21.996 --> 0:03:26.836 possible to control with insulin and other medications, and that 0:03:26.996 --> 0:03:30.636 just I felt really helpless. I felt helpless, and I 0:03:30.756 --> 0:03:35.116 felt this urge too. Okay, I don't feel comfortable going 0:03:35.156 --> 0:03:39.316 down this intellectual curiosity path of becoming a physicist, and like, 0:03:39.396 --> 0:03:42.356 what can I do to build better tools? There must 0:03:42.356 --> 0:03:46.476 be better tools that are necessary if my grandma died 0:03:47.156 --> 0:03:49.836 pretty odd, you know, due to diabetes, and there was 0:03:49.876 --> 0:03:52.396 nothing anyone could really do about it. 0:03:52.916 --> 0:03:55.316 So she started thinking about how to use physics to 0:03:55.436 --> 0:03:58.916 improve human health. For her graduate work, she figured out 0:03:58.916 --> 0:04:02.716 how to use lasers to make tiny holes in cell walls, 0:04:03.396 --> 0:04:06.236 and she started talking to all the researchers that she 0:04:06.276 --> 0:04:11.356 could to try and find useful applications for her research. Eventually, 0:04:11.476 --> 0:04:15.356 someone told her about induced pluralpotent stem cells, about iPSCs, 0:04:15.916 --> 0:04:17.676 and she realized that she might be able to use 0:04:17.756 --> 0:04:21.716 lasers to help automate the process of cultivating iPSCs. 0:04:22.076 --> 0:04:25.716 IPCs are extra complicated and special. I thought, I like 0:04:25.756 --> 0:04:27.956 to call them special because you actually have to go 0:04:28.076 --> 0:04:30.996 and like scrape bad cells away with a pipe pedant. 0:04:31.396 --> 0:04:32.596 It's super artisanal. 0:04:32.716 --> 0:04:36.076 So you're basically having these brilliant scientists looking under a microscope, 0:04:36.636 --> 0:04:39.636 holding cells in the dish, and then scraping with a 0:04:39.676 --> 0:04:44.316 pipe petter. And they're literally working ten ten hours a 0:04:44.396 --> 0:04:47.756 day scraping cells, not taking vacations, trying to get to 0:04:47.796 --> 0:04:49.956 work during the craziest snowstorms. 0:04:49.396 --> 0:04:52.116 Because if they don't show up, that run dies. 0:04:52.756 --> 0:04:55.516 Uh huh. And presumably, I mean if you think of 0:04:55.636 --> 0:04:58.036 actually getting it to the point where you can treat patients, 0:04:58.876 --> 0:05:03.436 it would be sort of impossibly expensive slash small scale, right, 0:05:03.516 --> 0:05:06.996 Like you could never do that for one hundred thousand 0:05:07.036 --> 0:05:08.476 people or something, right. 0:05:09.236 --> 0:05:11.356 Heart, we don't have enough scientists in the world. 0:05:11.756 --> 0:05:15.076 And of course that like highly skilled labor means like 0:05:15.276 --> 0:05:17.756 even more expensive than expensive drugs usually are. 0:05:17.836 --> 0:05:19.316 Right. Presumably that's right. 0:05:19.396 --> 0:05:22.836 And I think the cost estimates have gotten down over 0:05:22.876 --> 0:05:24.956 the past couple years because people have figured out how 0:05:24.996 --> 0:05:27.996 to do better biology but we're still in the hundreds 0:05:27.996 --> 0:05:29.196 of thousands of dollars. 0:05:29.876 --> 0:05:33.356 So you're setting out to solve this problem of how 0:05:33.396 --> 0:05:37.236 to grow iPSCs at scale, and in particular right this 0:05:37.396 --> 0:05:40.356 problem of getting rid of the bad cell colonies without 0:05:40.356 --> 0:05:43.396 disturbing the good ones. How do you figure out how 0:05:43.436 --> 0:05:43.836 to do that? 0:05:44.596 --> 0:05:46.196 So one of the things we did at the time 0:05:46.316 --> 0:05:49.356 was we spent quite a bit of time with biologists, 0:05:49.756 --> 0:05:51.916 so we were trying to understand their workflow. What do 0:05:51.996 --> 0:05:53.396 they do, what do they do in the lab, what 0:05:53.436 --> 0:05:54.436 are they looking for? 0:05:54.556 --> 0:05:59.036 How are they doing their hand gestures? And there's a 0:05:59.036 --> 0:06:01.196 lot of tilting, there's like tapping. 0:06:01.636 --> 0:06:05.516 This is the artisanal sort of separating the sales process exactly. 0:06:05.556 --> 0:06:08.476 The entire IPS manufacturing process can be on the order 0:06:08.516 --> 0:06:10.916 of three to four and there are many different parts 0:06:10.956 --> 0:06:14.196 to it. And certain scientists have certain features that they 0:06:14.236 --> 0:06:17.316 see by eye. So some will describe smiley faces, some 0:06:17.396 --> 0:06:19.756 will describe other features that they're seeing. 0:06:19.516 --> 0:06:22.836 Meaning like they look through the microscope to distinguish the 0:06:22.876 --> 0:06:24.956 good from the bad. That's right, and they sort of 0:06:24.996 --> 0:06:25.836 know it when they see it. 0:06:25.996 --> 0:06:26.396 That's right. 0:06:26.476 --> 0:06:28.916 And you have a whole range of how good our 0:06:28.996 --> 0:06:33.076 scientists are globally. So the best scientists are sitting at 0:06:33.876 --> 0:06:36.316 inside cleanrooms looking at these cells and trying to decipher 0:06:36.356 --> 0:06:39.116 the best cells. And the stakes are high because you 0:06:39.116 --> 0:06:41.596 don't really get in the way they're manufacturing their cells. 0:06:42.196 --> 0:06:45.076 They don't get to necessarily test the cells until you've 0:06:45.236 --> 0:06:46.396 done the entire process. 0:06:46.476 --> 0:06:48.556 Yeah. So, just to be clear, like this is a 0:06:48.636 --> 0:06:52.396 disaster for drug manufacturing, right, like not even for safety, 0:06:52.436 --> 0:06:55.716 but just like it's never going to work, right, It's 0:06:55.756 --> 0:06:57.996 never gonna work at scalesh Sure it could work for 0:06:58.036 --> 0:07:00.676 research for a while, but like you or somebody like 0:07:00.716 --> 0:07:02.796 you needs to come along to automate this, right. 0:07:03.396 --> 0:07:06.556 I'm so glad I never thought about how hard it 0:07:06.596 --> 0:07:09.316 was going to be and just went for it. If 0:07:09.356 --> 0:07:12.076 I know how hard it was going to be, like 0:07:12.116 --> 0:07:15.476 startups are hard, but I think we had this curiosity. 0:07:15.756 --> 0:07:18.636 We're a bunch of physicists, We had energy, we were passionate, 0:07:19.316 --> 0:07:21.196 and my co founder Marine and I we are very 0:07:21.236 --> 0:07:24.876 passionate about being in medicine and using our physics knowledge 0:07:25.316 --> 0:07:27.956 for medicine. So we were like, Okay, why don't we 0:07:27.996 --> 0:07:29.916 try why don't we figure out how to do it? 0:07:29.956 --> 0:07:35.396 And between twenty twenty and now, so many things have 0:07:35.516 --> 0:07:39.156 happened on the science and technology front at Seleno that 0:07:39.276 --> 0:07:41.436 I would label as impossible. 0:07:42.236 --> 0:07:44.956 I think there were at least the first few years. 0:07:44.756 --> 0:07:46.196 Of that phase, I was like, I don't know if 0:07:46.236 --> 0:07:49.316 any of this will work, and it was hard for 0:07:49.356 --> 0:07:52.356 me to come to terms with given that we had 0:07:52.556 --> 0:07:53.516 raised a big round. 0:07:53.596 --> 0:07:55.356 You know, a lot of people were talking about. 0:07:55.196 --> 0:07:58.316 This, but I also felt I kept telling my team, 0:07:58.356 --> 0:08:00.516 it's like, we have to give it our best shot, 0:08:00.796 --> 0:08:05.796 because if a team like ours isn't brave enough to 0:08:05.916 --> 0:08:08.316 try to go after this, this might not be resolved 0:08:08.676 --> 0:08:11.916 for a few decades. And what that means is we 0:08:12.476 --> 0:08:16.516 are at risk of falling into this pattern of pushing 0:08:16.596 --> 0:08:21.236 through complex manufacturing, of selling gene therapy products, even getting 0:08:21.276 --> 0:08:24.716 them through a phase through approval, but once they hit commercial, 0:08:25.276 --> 0:08:27.996 they're not meeting the patients. 0:08:28.516 --> 0:08:31.916 Meaning even if it works with the kind of technology 0:08:31.916 --> 0:08:34.836 that existed before you came along, even if it works therapeutically, 0:08:35.356 --> 0:08:38.836 it'll be so sort of bespoke and expensive that most 0:08:38.876 --> 0:08:40.476 people in the world aren't going to get it even 0:08:40.516 --> 0:08:41.956 if they need it. Is that what that means? 0:08:43.116 --> 0:08:45.996 That is what that means, And you know iPSCs, it's 0:08:46.036 --> 0:08:49.396 still to be determined. We don't have any first approval, 0:08:49.476 --> 0:08:51.476 so let's see how that goes. But it is the 0:08:51.516 --> 0:08:56.036 most complex manufacturing process that I've seen so far. Other areas, 0:08:56.076 --> 0:08:59.356 other examples I can point to are cart therapies, Cancer 0:08:59.396 --> 0:09:01.236 therapy is curative, incredible. 0:09:01.676 --> 0:09:04.036 This is another cell therapy and the domain of like 0:09:04.116 --> 0:09:06.556 let's take cells and develop them and give them to 0:09:06.596 --> 0:09:10.996 the patient. Right therapy is the sort of the signal 0:09:11.036 --> 0:09:12.596 achievement of self therapy so far. 0:09:12.756 --> 0:09:13.516 Right, it was. 0:09:13.556 --> 0:09:17.436 Huge, and that was happening that approval was coming up 0:09:18.516 --> 0:09:21.756 when I was starting Seleno, so it actually was very 0:09:21.796 --> 0:09:24.716 motivating to see, Wow, now we can actually reach the 0:09:24.796 --> 0:09:27.756 point of curing previously in kurbal disease. But it's been 0:09:27.796 --> 0:09:32.156 interesting to see about the cart space. Every year, the 0:09:32.196 --> 0:09:34.716 maximum number of patients being dosed with a CARTI on 0:09:34.716 --> 0:09:38.436 the year is on average still ten thousand patients annually. 0:09:38.556 --> 0:09:40.836 Which is a small number. Which is a small number, 0:09:40.996 --> 0:09:43.476 what hundreds of thousands or millions who might benefit from 0:09:43.476 --> 0:09:44.636 the tree correct. 0:09:44.316 --> 0:09:47.916 And now the scope is expanding to solid tumors and autommunity, 0:09:47.996 --> 0:09:51.116 So like, why are we stuck? We're stuck in this number. 0:09:51.116 --> 0:09:52.556 There's an infrastructure problem. 0:09:52.636 --> 0:09:56.156 It's a scale problem. It's a lack of manufacturing at scale. 0:09:56.476 --> 0:10:00.996 That's correct, and so I do feel very strongly that 0:10:01.036 --> 0:10:06.356 it's important to push forward trials and build for scale 0:10:06.516 --> 0:10:09.916 from the get go. But it's hard because it's tempting 0:10:09.956 --> 0:10:12.036 to take shortcuts and like, oh, we could do this 0:10:12.156 --> 0:10:14.636 and this will be much faster. And the question I'm 0:10:14.636 --> 0:10:17.756 always asking my team is Okay, great, so let's take 0:10:17.756 --> 0:10:20.916 a step back, but how will this solution? Could this 0:10:21.036 --> 0:10:23.476 address a million patients annually? 0:10:23.636 --> 0:10:25.876 Yeah, because there are a bunch of sort of cluges 0:10:25.876 --> 0:10:27.836 where you could make a thing that works and it's 0:10:27.876 --> 0:10:30.996 more efficient than what people do today, but it's still 0:10:31.116 --> 0:10:33.636 kind of clue g and artisanal and not great to scale. Like, 0:10:33.676 --> 0:10:34.596 that's the trade off. 0:10:35.116 --> 0:10:37.676 That is the trade off, and that's an everyday trade off. 0:10:37.676 --> 0:10:40.316 And I feel very blessed that we've been at it 0:10:40.396 --> 0:10:43.436 for quite some time, building the different layers, and we're 0:10:43.476 --> 0:10:46.076 at this point now where we're building our closed system. 0:10:46.156 --> 0:10:51.636 It's called Nebula, Okay, and it's it's it's the next version. 0:10:51.676 --> 0:10:54.356 It's the version I never imagined even when we were 0:10:54.356 --> 0:10:58.516 building the laser based technology, which we call the optical bioprocess. 0:10:59.396 --> 0:11:03.276 But the idea is in order to ultimately scale down cost. 0:11:04.596 --> 0:11:08.836 You know, it's always great to run high precision manufacturing, 0:11:09.636 --> 0:11:13.476 so we have that down, but the cleanroom costs are still. 0:11:13.276 --> 0:11:17.516 High because again you're dealing with living cells and they 0:11:17.556 --> 0:11:20.396 can't get contaminated because that would be disastrous. So you 0:11:20.476 --> 0:11:23.036 have to have a crazy super sterile environment. 0:11:23.796 --> 0:11:27.756 And we're doing many patients, and you want to parallelize 0:11:27.756 --> 0:11:29.796 as much as possible, and you want to protect all 0:11:29.836 --> 0:11:34.276 the patient samples from each other. So what we're building now, 0:11:34.276 --> 0:11:38.876 which is even harder but exciting and we're making some 0:11:38.916 --> 0:11:43.796 great progress, is the closed version of this so building 0:11:43.916 --> 0:11:49.396 cassettes that could ultimately be the clean space that all 0:11:49.436 --> 0:11:52.316 the manufacturing happens. Is the size of your iPhone, which 0:11:52.356 --> 0:11:53.756 is very exciting, huh. 0:11:53.796 --> 0:11:55.996 So it's basically, the size of your iPhone is a 0:11:56.076 --> 0:12:00.876 tiny clean room where you are cultivating induced player potent 0:12:00.916 --> 0:12:04.116 stem cells. It all happens inside that little box. 0:12:04.716 --> 0:12:05.876 That is what's being designed. 0:12:05.916 --> 0:12:08.876 I just want to put lots of little stars next 0:12:08.916 --> 0:12:11.756 to it, because this is all in collaboration with the FDA. 0:12:12.156 --> 0:12:15.276 You know, the type of vision we're portraying is not 0:12:15.476 --> 0:12:18.916 how cell therapy manufacturing is happening today. So we're gonna 0:12:18.916 --> 0:12:21.076 work with the FDA and we're going to present data 0:12:21.116 --> 0:12:23.596 and we're going to work with our clinical collaborators. But yes, 0:12:23.636 --> 0:12:26.876 wouldn't that be awesome because then we can we have 0:12:27.076 --> 0:12:31.276 these autonomous systems with close cassettes. We don't need to 0:12:31.396 --> 0:12:33.636 establish high grade cleaner So you get a ton of 0:12:33.636 --> 0:12:38.156 flexibility on setting up manufacturing in places where you might 0:12:38.156 --> 0:12:40.876 not have clean room, you might not have academic centers 0:12:40.876 --> 0:12:42.636 of excellent, you might not have the workforce. 0:12:42.916 --> 0:12:46.076 Yeah, Like the dream is it happens inside a machine essentially, 0:12:46.156 --> 0:12:49.076 like you buy the machine and all the artisanal knowledge 0:12:49.076 --> 0:12:51.156 and the clean room and everything is in the machine 0:12:51.196 --> 0:12:52.076 is embedded. 0:12:51.676 --> 0:12:55.596 In the machine, intelligent machines that make your best them cells. 0:12:55.716 --> 0:12:59.676 Yep. So that so we've talked about the past, and 0:12:59.676 --> 0:13:01.436 now we've talked about the dream for the future. Let's 0:13:01.436 --> 0:13:03.436 talk about the present. I read Is it right that 0:13:03.476 --> 0:13:08.516 there's a Phase one trial in Parkinson's that you're involved in? 0:13:08.556 --> 0:13:09.636 What's with that? 0:13:10.036 --> 0:13:14.556 Yes, we had a very exciting year. We found our 0:13:14.636 --> 0:13:21.356 first clinical collaborator. It's the Parkinson's Cell Therapy team at 0:13:21.436 --> 0:13:26.236 mess General Brigham and it's been great to work with them. 0:13:26.276 --> 0:13:29.956 They have a phase one running where they're making the 0:13:30.036 --> 0:13:34.956 patient's own IPSS and their own neurons, and these neurons 0:13:34.956 --> 0:13:38.836 are being transplanted into the brain, okay, and we're working 0:13:38.836 --> 0:13:42.196 together to do the transfer of the manufacturing into our 0:13:42.276 --> 0:13:44.636 automated platform. And there's a lot of back and forth 0:13:44.716 --> 0:13:47.956 right now on getting that collaboration up and running. 0:13:48.156 --> 0:13:51.636 And so just to be clear, that trial exists now 0:13:51.676 --> 0:13:55.076 and they're currently doing it in the artisanal, old fashioned, 0:13:55.596 --> 0:13:58.396 hard and expensive way, and the hope is that you 0:13:58.436 --> 0:14:03.276 will come in and do it in your faster, cheaper way. 0:14:04.036 --> 0:14:05.036 That's exactly right. 0:14:05.316 --> 0:14:10.956 They're incredibly brave scientists and clinicians and they've put in 0:14:10.956 --> 0:14:14.356 the hard work over the past few decades to get 0:14:14.396 --> 0:14:19.636 this bioprocess up and running, invent new surgical techniques, design 0:14:19.676 --> 0:14:23.396 the clinical trial, and yeah, we want to support them 0:14:23.436 --> 0:14:26.396 to scale through the trial and then how does it 0:14:26.396 --> 0:14:30.036 get into commercial So it's been really exciting. And one 0:14:30.076 --> 0:14:32.076 of the things I love about this collaboration is we're 0:14:32.116 --> 0:14:36.756 geographically very close because we're in Cambridge and they're across 0:14:36.836 --> 0:14:39.276 the River Master. I think it's like a mile or 0:14:39.316 --> 0:14:44.556 two from us. So the ability to connect dots has 0:14:44.556 --> 0:14:49.156 been incredible because the manufacturing is happening inside the hospital. 0:14:50.196 --> 0:14:52.956 Interesting, like you're building a machine in the hospital. 0:14:53.116 --> 0:14:55.476 We will be deploying the machine in the hospital. Yes, 0:14:55.796 --> 0:14:57.956 the machines are being built at our headquarters. 0:14:57.956 --> 0:15:00.516 Now I shouldn't say built. Yeah, so you're going to 0:15:01.036 --> 0:15:03.356 drive put the machine in a truck. How big is 0:15:03.396 --> 0:15:05.356 the machine by the way, you know. 0:15:06.636 --> 0:15:10.956 Maybe like one or two refrigerators. Yeah, so that's how 0:15:10.956 --> 0:15:12.596 big they are. And the cassettes are about the size 0:15:12.596 --> 0:15:17.316 your iPhone. And initially we'll be deploying the machines to 0:15:17.356 --> 0:15:19.716 be doing making sales for one patient at a time, 0:15:19.836 --> 0:15:22.676 and then as we build more evidence, we'd love to 0:15:22.716 --> 0:15:24.916 make more and more patients on each machine. 0:15:25.116 --> 0:15:30.196 And will that be the first time your sales are 0:15:30.236 --> 0:15:31.956 going into patients. 0:15:31.676 --> 0:15:32.396 Yes, I believe. 0:15:32.476 --> 0:15:32.556 So. 0:15:33.156 --> 0:15:37.236 We do have two other amazing collaborators, one in South Korea. 0:15:37.276 --> 0:15:39.636 They're working on a peripheral artery disease. They also have 0:15:39.716 --> 0:15:43.076 a phase one trial. They're going to have a US 0:15:43.876 --> 0:15:48.116 expansion and then a spinal cord injury company as well. 0:15:48.116 --> 0:15:52.036 But yeah, the mass General team is doing great. We're 0:15:52.076 --> 0:15:58.476 also leaning into collaborating with the FDA We've had a 0:15:58.596 --> 0:16:02.156 very positive experience working with them. Earlier this year we 0:16:02.156 --> 0:16:07.436 were granted our Advanced Manufacturing Technology designation for iPSC generation 0:16:08.476 --> 0:16:11.436 and it's I think that maybe that's the biggest surprise 0:16:11.476 --> 0:16:15.836 of the year at how technology forward the FDA is 0:16:15.876 --> 0:16:18.716 and they've been paying attention. They've been asking great questions 0:16:18.716 --> 0:16:21.156 and all of our meetings. They have a full stack 0:16:21.276 --> 0:16:26.156 AI team that is fully tuned into how our models 0:16:26.156 --> 0:16:29.756 are being trained, like what's what's behind the hood. So 0:16:29.796 --> 0:16:32.756 it's really really compelling. Yeah, because I think the last 0:16:32.796 --> 0:16:36.036 ten years of selling gene therapies have been transformative in 0:16:36.116 --> 0:16:40.796 terms of curative medicines, but everybody is missing the impact of. 0:16:40.756 --> 0:16:42.516 Scale, like including the regulator. 0:16:42.556 --> 0:16:46.916 So they've now taken upon themselves to sort of help 0:16:47.476 --> 0:16:50.956 the field think about scale and work with technology companies 0:16:51.036 --> 0:16:56.916 like ourselves. So it's incredibly it's special and. 0:16:56.996 --> 0:17:01.396 Just briefly, like, from say the patient's point of view, 0:17:01.876 --> 0:17:03.516 how will it work? Simply? 0:17:04.276 --> 0:17:06.516 Yeah, I mean you know, we can take the mass 0:17:06.516 --> 0:17:07.716 General Brigham. 0:17:07.396 --> 0:17:09.076 Yeah, in that case, what will happen? 0:17:09.756 --> 0:17:13.076 In that case, they'll be they'll see a physician and 0:17:13.116 --> 0:17:16.236 they'll get a diagnosis for their disease. And in this 0:17:16.316 --> 0:17:20.476 case it's Parkinson's. Then they'll get a prescription that says 0:17:21.516 --> 0:17:25.956 you're going to get an atologous cell therapy. 0:17:25.516 --> 0:17:27.676 And meaning your own cells. 0:17:27.796 --> 0:17:32.076 Yes, yes, exactly, personalize your own match with your DNA. 0:17:32.396 --> 0:17:35.516 And then they'll show up to or maybe it's the 0:17:35.516 --> 0:17:37.556 same day they get the prescription and maybe they're at 0:17:37.556 --> 0:17:42.596 their doctor's office, but they will have to either provide 0:17:43.116 --> 0:17:47.516 blood draw or a skin biopsy. I can imagine really 0:17:47.556 --> 0:17:49.556 far into the future. It could be hair cells, it 0:17:49.556 --> 0:17:50.476 could be saliva. 0:17:50.676 --> 0:17:53.356 But for this one, yeah, they take a little bit 0:17:53.356 --> 0:17:54.996 of your blood or a little bit of your skin 0:17:55.116 --> 0:17:55.716 and then. 0:17:56.636 --> 0:18:00.076 And then they say, okay, we'll schedule your surgery and 0:18:01.196 --> 0:18:03.036 come back in a couple months and. 0:18:02.996 --> 0:18:03.756 We'll be ready to go. 0:18:04.236 --> 0:18:07.276 Essentially, you're starting with the patient's cells, their skin cells 0:18:07.316 --> 0:18:10.916 or their blood cells. You're ending up with brain cells 0:18:10.956 --> 0:18:13.716 that match their own brain cells. What part of that 0:18:13.756 --> 0:18:16.396 transition happens inside your machine automatically. 0:18:16.716 --> 0:18:23.076 Yes, So the getting to really high quality iPSCs is 0:18:23.116 --> 0:18:25.516 the first product we've established. 0:18:25.516 --> 0:18:27.876 We're establishing end to end, Like. 0:18:27.796 --> 0:18:30.116 You put the whatever the blood cell or the skin 0:18:30.196 --> 0:18:32.236 sell in the machine and out the other end comes 0:18:32.236 --> 0:18:35.516 a high quality ips is it really like that. 0:18:37.036 --> 0:18:39.996 It's lots of AI and lots of biology and lots 0:18:40.036 --> 0:18:41.476 of fluidages, but it is kind of like that. 0:18:41.676 --> 0:18:44.916 Yeah, okay, it's true that part is automated. That's really 0:18:44.956 --> 0:18:46.116 what I'm asking, Like. 0:18:46.836 --> 0:18:49.156 We're working on it. Yeah, it's and it's going to 0:18:49.236 --> 0:18:53.396 be automated. There will be human human experts in loop 0:18:53.636 --> 0:18:56.716 as always, and there will be end QC. But the 0:18:56.796 --> 0:18:58.956 day to day operations, I mean, you know right now, 0:18:58.996 --> 0:19:03.636 like Exeleno, things here are running automated and they run 0:19:03.796 --> 0:19:06.516 pretty much twenty four hours a day, so they're like, 0:19:07.436 --> 0:19:13.956 you know, proming imaging, fluids, laser processing because the cells 0:19:14.076 --> 0:19:17.156 need different actions at different time points, so a lot 0:19:17.196 --> 0:19:20.236 of those things we've established to be algorithmic. 0:19:20.316 --> 0:19:21.436 Yeah, it is automated. 0:19:21.916 --> 0:19:24.836 You know, we have a we have a very small 0:19:24.876 --> 0:19:27.396 and mighty team of bio engineers, but we definitely do 0:19:27.476 --> 0:19:28.116 the work. 0:19:27.916 --> 0:19:31.596 Of ten fifty x more. 0:19:31.836 --> 0:19:34.476 I would say, yeah, through automation, through. 0:19:34.316 --> 0:19:35.436 Automation, that's right. 0:19:35.596 --> 0:19:46.196 Yeah, we'll be back in just a minute. 0:19:51.316 --> 0:19:53.756 One quick note before we get back to the interview. 0:19:54.356 --> 0:19:57.636 Near the end of the conversation, you will hear Nabiha 0:19:57.836 --> 0:20:04.276 mentioned something called allogeneic therapies. Those are therapies where ipsc's 0:20:04.876 --> 0:20:08.556 induced player potent stem cells are developed based on generic 0:20:08.636 --> 0:20:12.996 cells rather than based on a patient's own cells. Those 0:20:13.076 --> 0:20:15.716 are easier to cultivate, but in many cases they have 0:20:15.836 --> 0:20:20.236 drawbacks similar to organ transplants, because patient's own immune systems 0:20:20.276 --> 0:20:22.716 tend to reject those cells. So I just wanted to 0:20:22.796 --> 0:20:26.396 clarify that in advance. Okay, back to the interview. How 0:20:26.396 --> 0:20:30.516 are you using machine learning or AI in your automated process. 0:20:31.996 --> 0:20:33.996 The way we use machine learning and AI is we 0:20:34.036 --> 0:20:37.516 take a lot of photos of all the cells every day. 0:20:37.996 --> 0:20:38.836 How are they doing. 0:20:38.916 --> 0:20:41.756 We've trained a bunch of algorithms in Google Cloud that 0:20:41.916 --> 0:20:44.236 tell you, oh, this is a good stem cell, this 0:20:44.316 --> 0:20:46.196 is not a good stem cell, this is good density, 0:20:46.236 --> 0:20:50.916 this is not good density. And then those algorithms feed 0:20:50.956 --> 0:20:53.556 into other algorithms that make decisions on what to do 0:20:53.596 --> 0:20:56.836 with the cells. So that your expert scientists don't have 0:20:56.916 --> 0:20:59.276 to sit and make all these decisions, and it's hard 0:20:59.316 --> 0:21:01.996 to make when it's at this massive of a scale. 0:21:02.196 --> 0:21:05.076 So they can review what the algorithms are doing, they 0:21:05.116 --> 0:21:08.196 can intervene it whenever they want to. I think like now, 0:21:08.236 --> 0:21:11.516 a very timely a similar field would be self driving cars. 0:21:11.556 --> 0:21:14.756 You know, there's just a lot of imaging that's being used, 0:21:14.796 --> 0:21:17.396 and then the car can make its own decisions. And 0:21:17.716 --> 0:21:19.956 at least in a tesla you have you have a 0:21:20.036 --> 0:21:23.476 driver and they can override any time. Wimos are running autonomous. 0:21:24.196 --> 0:21:25.116 So that's what we do. 0:21:25.156 --> 0:21:28.076 We use imaging to help with all the decision making 0:21:28.156 --> 0:21:29.036 your manufacturing. 0:21:29.156 --> 0:21:32.436 So it's largely pattern matching, right, which is essentially what 0:21:32.876 --> 0:21:35.036 the expert scientists are doing. As you said, some of 0:21:35.076 --> 0:21:37.396 them talk about whatever a smiley faces something, it's just 0:21:37.436 --> 0:21:41.116 because the individuals have seen whatever thousands have done it 0:21:41.116 --> 0:21:44.476 thousands of times, and it's a it seems like a 0:21:44.516 --> 0:21:47.076 good thing to use AI for, right, like a classic like, 0:21:47.156 --> 0:21:49.436 here's lots and lots of good ones and lots of 0:21:49.476 --> 0:21:51.756 lots of bad ones. Now here's a new one. Pickwa 0:21:51.796 --> 0:21:52.996 are the good ones and one of the bad ones? 0:21:53.036 --> 0:21:55.956 It's that you're absolutely right. So what's great about it. 0:21:56.116 --> 0:21:59.036 Usually if humans can see something by eye, we're able 0:21:59.116 --> 0:22:00.636 to trend an algorithm to do that. 0:22:00.716 --> 0:22:01.756 So that's real number one. 0:22:01.836 --> 0:22:03.756 And then the second thing that's very exciting about what 0:22:03.796 --> 0:22:09.316 we do is this time series data. This is really 0:22:09.356 --> 0:22:12.836