Latest / StartUp Health NOW Podcast / AI-Powered Health: Naheed Kurji of Cyclica Takes a Deep Dive into Artificial Intelligence for the Healthcare Industry

Transcript

AI-Powered Health: Naheed Kurji of Cyclica Takes a Deep Dive into Artificial Intelligence for the Healthcare Industry · StartUp Health NOW Podcast

• 0:00
  [on-hold music] So instead of building the next algorithm to go narrow on one biological protein target and find chemistry for that, we said, "Let's look at all proteins that exist, some of which that are not really well understood, some of which don't have data around, some of which, you know, we can't work on classically, but let's build a model to evaluate the entire collection of proteins called the proteome."
• 0:29
  And we're the only company that has built a technology like that. [upbeat music] Welcome to Startup Health Now, the podcast where we celebrate the entrepreneurs and innovators transforming health.
• 0:40
  I'm your host, Logan Plaster. So here at Startup Health, our mission is to inspire, educate, and invest in health transformers.
• 0:48
  This is our term for those mission-driven entrepreneurs who are putting it all on the line to improve health.
• 0:54
  But the beauty of having such a rich portfolio spanning four hundred and thirty-plus investments across twenty-nine countries is that that inspiration and education flows both ways.
• 1:04
  My guest today, Naheed Kurji, is the CEO and co-founder of Cyclica, and he's gonna get us up to speed on the world of AI-powered drug discovery.
• 1:13
  We'll learn how artificial intelligence and machine learning is speeding up our ability to identify molecules and market-tested drugs for new use cases.
• 1:21
  Naheed will give us an inside look at how Cyclica is specifically trying to reshape this market, then he'll zoom way out.
• 1:28
  As the chair of the Alliance for Artificial Intelligence in Health, Naheed can help us understand where the AI for health industry is going at a macro level. Stick around.
• 1:36
  [upbeat music] Naheed, we've talked many times over the years, and something that I always appreciate about my conversations with you is that I know I'm gonna get an education.
• 1:47
  You lead a, uh, a brilliant team of scientists really at the tip of the spear of the applications of AI and machine learning to discover new molecules that will help us find new drugs, f- uh, pharmaceuticals and therapies.
• 2:01
  Uh, it's a field that, that I don't know well, and, uh, I wanna learn more from you about it today. I wanna start by asking you, how do you explain what Cyclica does and kinda what you're working on?
• 2:03
  Let's say you're at a cocktail party and you need to explain it to someone who's not in the AI drug discovery industry.
• 2:03
  [sighs] So instead of building the next algorithm to go narrow on one biological protein target and find chemistry for that, we said, "Let's look at all proteins that exist, some of which that are not really well understood, some of which don't have data around, some of which, you know, we can't work on classically, but let's build a model to evaluate the entire collection of proteins called the proteome."
• 2:04
  And we're the only company that has built a technology like that. Welcome to Startup Health Now, the podcast where we celebrate the entrepreneurs and innovators transforming health. I'm your host, Logan Plaster.
• 2:04
  So here at Startup Health, our mission is to inspire, educate, and invest in health transformers. This is our term for those mission-driven entrepreneurs who are putting it all on the line to improve health.
• 2:04
  But the beauty of having such a rich portfolio spanning four hundred and thirty-plus investments across twenty-nine countries is that that inspiration and education flows both ways.
• 2:04
  My guest today, Naheed Kurji, is the CEO and co-founder of Cyclica, and he's gonna get us up to speed on the world of AI-powered drug discovery.
• 2:04
  We'll learn how artificial intelligence and machine learning is speeding up our ability to identify molecules and market-tested drugs for new use cases.
• 2:04
  Naheed will give us an inside look at how Cyclica is specifically trying to reshape this market, then he'll zoom way out.
• 2:04
  As the chair of the Alliance for Artificial Intelligence in Health, Naheed can help us understand where the AI for health industry is going at a macro level. Stick around.
• 2:04
  [upbeat music] Naheed, we've talked many times over the years, and something that I always appreciate about my conversations with you is that I know I'm gonna get an education.
• 2:04
  You lead a, uh, a brilliant team of scientists really at the tip of the spear of the applications of AI and machine learning to discover new molecules that will help us find new drugs, f- uh, pharmaceuticals and therapies.
• 2:04
  Uh, it's a field that, that I don't know well, and, uh, I wanna learn more from you about it today. I wanna start by asking you, how do you explain what Cyclica does and kinda what you're working on?
• 2:13
  Let's say you're at a cocktail party and you need to explain it to someone who's not in the AI drug discovery industry. Yeah. Love the question. Um, the answer is fairly easy.
• 2:24
  Within there, there's some nuance that I think it gets a little bit complicated. We're all consumers of medicines, if you think about it. We're all patients in some way, shape, or form.
• 2:36
  Now, what we don't generally understand is where those medicines come from, how they were discovered, how they were de-developed, the complexity of that entire process. So let's start there.
• 2:48
  To go from a specific disease to a medicine for that disease is a highly onerous, costly, time-consuming endeavor. Um, and it on average, now people will debate the following, but reports have come out that on average
• 3:07
  it takes over a decade and costs billions of dollars, hundreds of millions up to billions of dollars to get that first medicine to a patient.
• 3:16
  And if you break up that time and that cost, there's really two steps to bringing a medicine to a patient. There's the discovery step, step one, and then there's the development step, step two.
• 3:29
  Within each, there's many sub-steps. The discovery stage, you can think of that as everything pre-human, before a clinical trial. And within the development stage, there are three steps generally.
• 3:45
  Um, phase one clinical trials is what we hear about. That's generally on evaluating is this molecule safe? There is phase two clinical trials. That's is it efficacious?
• 3:56
  Will it have the therapeutic efficacy that is desirable? And then there's phase three, which is a much broader kinda combined safety and efficacy on a bigger population.
• 4:07
  Now, to get to touching a human, the regulatory bodies want to make sure that these molecules are generally evaluated as safe and efficacious before touching a human.
• 4:18
  And that's in basic research, and that's in discovery.
• 4:22
  Now, if you break up discovery and development into timeframes, it's about seven years for development, phase one, two, and three, about five to seven years in discovery.
• 4:31
  Now, those numbers change depending on the disease. For very complex diseases like Alzheimer's disease, it could be much longer. For more rare diseases where there's fast track offer, it could be much shorter.
• 4:43
  But on average, five years to seven years for discovery, seven years for development. Now, within discovery, which is where Cyclica spends a lot of its time,
• 4:53
  uh, there's a bunch of basic research that we feel is just inefficiently done, and it's done classically. It's done experimentally. You know, we--
• 5:03
  Um, many of the listeners of this podcast probably remember their chemistry and biology days in high school or university where you'd use your reagents and pipette a bunch of stuff and do a bunch of testing, look at some data, do something new.
• 5:14
  And that entire cycle was, we believe, fraught with opportunity to create a more scalable approach and bring technology based on the wealth of data that exists. And so back in twenty twelve, twenty thirteen timeframe,
• 5:31
  a wave of innovators came out to leverage the big data and analytical tools to evaluate that data to make, um, more strong predictions that could simulate what would otherwise be done in a lab setting.
• 5:48
  And there's a fairly seminal paper written by McKinsey & Company, uh, co-authored by a few folks, including Sastry Chilukuri, um, who's now the co-CEO of Medidata, a remarkable company that is doing a lot in the clinical trial space, a dear friend and one of Cyclica's board members.
• 6:06
  And the paper, if I recall specifically, it was titled How Big Data and Predictive AnalyticsCan or will revolutionize the pharmaceutical industry.
• 6:14
  And that paper and a few other activities that were going alongside that opened the doors to now what are hundreds of companies in this space. Now, if I t-- c-come back to your question on the back of that setting,
• 6:29
  the discovery of medicines is really complicated. You need to understand the disease, then you need to understand its biological driver of a disease, and generally, the biological drivers of a disease are proteins.
• 6:41
  When a protein is malfunctioning or doing something that it should not be doing, generally a disease is some down-- some way downstream going to happen.
• 6:50
  And then you need to find some form of chemistry or therapeutic to interact with that protein to stop it doing what it's doing that is malfunctioning.
• 6:58
  So you can think of the protein as the lock and the chemistry is the key, and the goal of drug discovery, simply put, this is very crass, and scientists will probably, you know, say that it's much more complicated, but if I bring it back simply, it's to find chemistry for a
• 7:15
  protein, a key for a lock. Mm-hmm.
• 7:18
  And that was generally done through very empirical methods, but an opportunity presented itself on the back of that paper, and a bunch of the data that has since been generated from the Human Genome Project, um, and a bunch of other efforts, is how can we do that computationally?
• 7:33
  So I'll pause, but that's the background, and I can talk a little bit more about what Cyclica does there. Now, just to be fair, I did ask you how you would explain it at a cocktail party- That's true...
• 7:43
  to someone who is unfamiliar with the market. So, uh, uh, I don't know what kind of cocktail parties you go to. How would you explain it in a couple of sentences?
• 7:52
  We apply big data predictive analytics tools like cloud computing to streamline the way in which medicines are discovered to get it out of a lab setting and bring it into a cloud-based setting. That's what I would say.
• 8:06
  I- But the cocktail parties that I go to- I love it... people are like, "Tell me more." [laughs] Exactly, exactly. Uh, help us understand briefly the landscape of AI drug discovered molecules and pharmaceuticals.
• 8:20
  Uh, if you're not in the industry, you might lump them all together, but of course, you understand there's this whole-- an entire gamut. So where does Cyclica fit in, in the market? Yeah.
• 8:28
  There are some remarkable companies that have built technologies to solve specific problems. So there's two types of-- generally two types of companies and two types of problems.
• 8:38
  The first type of company has existed for the past thirty to forty years, and these are companies that have built, um, physics-based computational methods to simulate how a drug, um, chemistry will interact with a biological protein target.
• 8:54
  But they do this with really deep resolution and bring, you know, how is this chemistry atomistically going to interact with this protein? And they do this in a simulated setting.
• 9:04
  And those companies have opened the door to a wave of computational methods to simulate, to make predictions of what otherwise is done in a lab setting. Now, there are strong limitations of the physics-based approach.
• 9:19
  Don't wanna get into all those today, but those are the first companies back in the late eighties, early, uh, nineties that entered into computational drug discovery.
• 9:28
  In the early to mid-2000s, with a bunch of data that started to gen-- be originated, generated, made available, post-Human Genome Project, a bunch of literature that started to come out, a bunch of companies started to emerge saying, "Let's take this data and make predictions on the back of that data.
• 9:44
  Let's not look at the atom-to-atom interaction between chemistry and biology, but let's parse literature and s-- um, make evaluations, predictions on the back of our interpretation or our model's interpretation of that literature."
• 9:58
  There are thousands of journals that come out every single year, millions that exist. It's too much for any one human or a group of people to do, so these models started to come out.
• 10:08
  Now, generally, the non-physics-based companies, the knowledge-based companies are the AI companies. The physics-based companies are generally non-AI in, in nature.
• 10:19
  Um, now the problems that these companies are trying to solve are, one, around how to find that key for that lock.
• 10:28
  The problem that Cyclica is solving is that we know as patients, drugs do more than one thing when they enter a biological system.
• 10:35
  It's not enough to find a key for a lock when you know a key is interacting with hundreds of locks.
• 10:40
  So instead of building the next algorithm to go narrow on one biological protein target and find chemistry for that, we said, "Let's look at all proteins that exist, some of which that are not really well understood, some of which don't have data around, some of which, you know, we can't work on classically, but let's build a model to evaluate the entire collection of proteins called the proteome."
• 11:05
  Yeah. And we're the only company that has built a technology like that. Interesting. Let's talk a little bit about Cyclica's journey and your journey. Uh, how long have you been, uh, at the CEO seat?
• 11:17
  Coming up to seven years in April. Um, the company's, um, older than that, but I took over the CEO April 12th of 2016. Tell me about that journey.
• 11:27
  I mean, I know you closed a Series B, a significant Series B a couple years ago. Just what has that business journey been like? Um, was it accelerated by COVID and sort of the decentralization of, uh, research?
• 11:39
  So what's it, what's it looked like? Yeah.
• 11:41
  So we closed our Series B, uh, we announced it June of twenty twenty, led by Drive Capital, supported by Chiesi Pharmaceuticals, you know, one of the big supporters of startup health, and it was based on an introduction from, uh, Katya Unity to the folks at Chiesi that led to their investment in us.
• 11:58
  And, you know, thank you continuously to you all. Um, you know, the journey from Cyclica to where we are today has been kind of on two paths.
• 12:07
  One on scientific and technical innovation, and one on business model innovation.You know, Cyclica is older than how long I've been a CEO, as I just mentioned.
• 12:16
  I co-founded the company as CFO and led kind of the back office, built the operations, did a lot of the business development in the very early days.
• 12:25
  But credit goes to where credit's due, kind of on the scientific ingenuity.
• 12:29
  My co-founder who brought me on to the company, Jason Medekidis, had this idea from well before he brought me on and started building a small team and a business model, uh, or a business plan around what Cyclica could do in the market.
• 12:42
  And the initial scientific innovation was, um, to take a chemi- a chemical compound and look at all the proteins that it could interact with and do something that nobody else could do.
• 12:52
  And we built a technology and a platform that we then sold to the pharma industry under a software as a service model.
• 13:00
  And so we had this really elegantly designed platform, could be accessed, could be used, molecules could be run. It was really nicely intuitive for scientists to access on their desktop.
• 13:12
  The business model was pay to access, pay to play, we'll generate revenue.
• 13:16
  And what we realized, Logan, was important for the future of the company and what we believe is the future of this industry, is that the value for a company like us that has built a technology that can generate novel insights, the value isn't from revenue from pharma.
• 13:32
  Pharma increasingly is outsourcing its drug discovery.
• 13:36
  Seventy-five percent of the discovery of medicines happens outside of the four walls of a pharma company, and that's an increase from fifty percent in the early 2000s.
• 13:45
  So the trend is that biotech companies are doing more of that drug discovery, but we as a company, we're servicing the pharma industry. So it took us some time.
• 13:55
  We realized that SaaS was not the right business model for us. Revenue was not the right metric for us.
• 14:01
  And once we started to see that our platform was driving really interesting insights, we turned it in on itself and we said, "Let's become a neo-biotech company on the back of our platform."
• 14:14
  And what I mean by that is using our AI augmented platform, let's create,
• 14:20
  advance, and own the largest portfolio of drug discovery programs in the industry and create a pipeline of assets that we can go back and sell to pharma companies because we know that they're relying on biotech companies to do that anyways.
• 14:33
  Mm-hmm. So instead of them evaluating hundreds of biotech companies, let's create a pipeline, a massive pipeline for them to evaluate with one company. COVID helped accelerate that. Oh yeah, go ahead please.
• 14:44
  Is, is, is that an ecosystem that all falls under the Cycl- Cyclica umbrella, uh, really clearly, or are those partnerships that you are, that you're creating around the world? Yeah, combination of both.
• 14:55
  I think it started off, you know, if I think of a teeter-totter, we had very few internal programs, very heav- heavily on partnerships. We've created a few joint ventures.
• 15:03
  One of the famous ones, and Startup Health brought us up on stage at the Startup Health Festival in two thousand and twenty, was, uh, a joint venture with Atai Life Sciences, a company called Entheogenics Biosciences.
• 15:15
  Um, the chemistry within Entheogenics is all from Cyclica, but Atai has the biological understanding of the mental health and neuropsychiatric diseases that we're going after, and they also have under their Atai umbrella, a deep expertise in like psychedelics as a medicine.
• 15:32
  Um, we came at it with the chemistry to do something really interesting to get a FDA approved molecule, hopefully, that recapitulate the therapeutic benefits of psychedelics for mental health disease.
• 15:45
  But that's a joint venture. We don't control the process. We just have an equity share of the company and the upside.
• 15:52
  As we have grown and raised more money, we have taken more control of some of our programs, so we've built an internal portfolio.
• 15:59
  But I'd say it's fifty-fifty right now, and increasingly as we grow, we'll raise more money, we'll advance some of our programs, and we'll take more control.
• 16:07
  So I think we'll be less reliant on partnerships as we have been historically, but right now it's probably about fifty-fifty. Got it.
• 16:16
  Uh, shifting gears, I mean, there's so much more we could dig into with what you've just said. Uh, there's a lot, a lot of depth there, but I'm gonna try to cover some more topics.
• 16:25
  I understand that you were recently named chair of the Alliance for Artificial Intelligence in Healthcare. Uh, so congratulations for that, but I wanna understand what does that entail?
• 16:38
  Uh, what are the issues that you're most intrigued by, uh, in those discussions with that, um, with that group? Yeah.
• 16:47
  The AAIH, the Alliance for Artificial Intelligence in Healthcare, was launched at the inaug- with the inaugural panel to launch the AIH was at J.P. Morgan Healthcare Conference two thousand and nineteen.
• 17:00
  Um, and the AIH was founded,
• 17:03
  I kid you not, back in February twenty eighteen, a few CEOs of hyper-innovative AI biotech companies, a few pharma execs that were at that time back in twenty eighteen when there was no AI in drug discovery development space, were coming together and saying, "There is a wave coming at us.
• 17:20
  Pharma's trying to evaluate. These guys are trying to innovate. We're trying to marry and we're trying to solve the same problems.
• 17:25
  Instead of us all solving those problems individually, what if we came together and tried to solve some of those problems collectively?"
• 17:32
  Yes, there are some competitive companies there and on the pharma side, but there are issues that we all believe that we could impact together. And so that was the basis of the AIH.
• 17:42
  It was co-founded by about thirteen companies, Cyclica being one. Um, and fast-forward to today, there's about fifty-five member organizations.
• 17:53
  Uh, once the organization was founded and launched, I took over one of the four executive officer positions as secretary, a role that I held for about three years until March of twenty twenty-two, uh, when the membership at our annual meeting appointed me as chair.
• 18:08
  And one of the things as chair, and Angelie Muller, um, from Roche as vice chair, we had a few re-- uh, a remarkable ex- executive officer group. What we're focused on, I'd say, are three big topics.
• 18:20
  Number one-We have a data working group. Within that data working group, we're talking about data standards, um, a really important issue, data access, data fairness.
• 18:32
  In the space of drug discovery and development, when you look at a bunch of the genetic and genomic data that exists in the world, about ninety to ninety-four percent of it, uh, comes from non-ethnic backgrounds, very European backgrounds.
• 18:46
  You know, while this is not woke or politically correct, I-- let's call a spade a spade. The data that comes from genetic genomics is very Caucasian lean driven.
• 18:56
  And so as a result, when we start to make predictions on personalized medicine, and you have data that you're training a model on that does not represent other populations, things fall off a cliff very quickly.
• 19:08
  So we talk a lot about data fairness, and next week, um, a few folks, including, you know, leader of AI at Microsoft, Amgen, uh, at Roche, and myself are leading a conversation that is sponsored by the AIH on the importance of solving the data bias issue in healthcare.
• 19:27
  The second thing that-- oh, so yeah, please go ahead. Yeah. Let me, let me pause you there and just say, what are some of the ideas on the table for tackling data fairness?
• 19:37
  I, I think first and foremost, it's a mindset issue. Um, the first thing is that people need to recognize that this is a problem that needs to be solved before bringing technology to solve it.
• 19:49
  I still don't believe, and I can say this as a diverse-- somebody of a diverse population, when I sit across from my peers, it's not the first thing that they think about. It's the first thing that I think about.
• 20:01
  It's the first thing that a few people that look and come from areas of the world that my ancestors come from think about. But the mindset isn't equal across the board.
• 20:10
  I've heard people in the genetics and genomics space that aggregate a lot of this data say, "We don't..." I've heard this, quite honestly.
• 20:19
  "It doesn't matter what that data looks like or where it comes from, so long as I can use it and do something with it." And I had to look at these people and like, "That makes no sense."
• 20:27
  Like, you have an influence to make sure that the data is the right data, that it's balanced, that it's representative, that it is fair. And so the mindset has to shift. Mm-hmm.
• 20:36
  Then-- a-a-and the impact is if you're training a model on just images of cats, and then you're asking that image-- that model to make a prediction on dogs, it's not gonna work.
• 20:49
  Not everybody looks like a ca-- and that's a very crass way of thinking about it, but if your data looks only like one set of things, it's gonna fail to make a representative prediction.
• 21:00
  Um, the other, I'd say, item is we certainly need to c-- we need to collect and share that data. So federated data is an important concept. Pharma companies sit on a lot of data on clinical trials.
• 21:12
  Hospitals have electronical medical records. There's a bunch of data, now privacy issues and security issues, but there is a wave of innovation to solve this through federated learning.
• 21:22
  There's a company in Canada called DNAStack that's doing some remarkable work. A Paris-based company called Oaken doing some great work there. So I think it starts with mindset. That's number one.
• 21:31
  Number two, I think there are some technology companies that are doing it, but we need to be doing much more as an industry. [lip smack] Interesting.
• 21:38
  When I think about the mindset piece of this, it makes me think about how you as a CEO manage a team of scientists.
• 21:46
  Here you are trying to, uh, sort of raise their gaze to these global issues when it's, it's very easy to be, um, sort of trapped in the data, uh, or just say, "Look, the numbers are the numbers. Uh, we're scientists.
• 22:00
  It is what it is." Uh, and you lead a team of, of brilliant scientists, but you have a, a financial background, a management background.
• 22:08
  Um, so I, I'm always curious talking to you to understand some of your techniques for, for managing this team, for, for, for guiding them, um, and balancing that sort of the science and the mindset. Yeah.
• 22:23
  It's a question that I ask myself every single day. Um, you know, I don't know how I've been so blessed to ha- maintain this seat at Cyclica. It's not-- it's my privilege to be in the seat of the CEO of Cyclica.
• 22:35
  It's not something I ever wanted, nor do I think I ever, you know, e-expected to have. And from time to time, I'm thinking, "Am I even the right person for this to s- help the company scale?"
• 22:44
  It's always in the top of my mind. So I don't think it's my right, and that's an important part of the mindset that I brought to this role, is that the company is more important than just me.
• 22:55
  And so what do I need to do as the dimmest bulb in the room? For those who don't know me that are listening to this, I don't have a PhD in computational biophysics or structural biology or medicinal chemistry.
• 23:07
  I don't have a master's of science in drug design. Um, but I say i-it's my passion, it's my life now, and I spend time asking a lot of questions.
• 23:17
  So I think there's three things that for me have helped me be successful in the role that I'm in. Number one, I don't know everything, and I'm totally fine with that.
• 23:26
  Number two, on the back of not knowing a bunch of stuff, I ask a lot of questions, and I frame very simple frameworks for me to then make a decision or guide a decision.
• 23:37
  And number three, uh, my entire objective is to bring the best people in the world to join Cyclica such that, and this is gonna sound crass, if I got hit by a bus today walking out this door and walking on the street, I get-- the company survives.
• 23:54
  It doesn't even just survive, it thrives, because we have the best people in the world who have come to Cyclica.
• 24:01
  You look at our leadership team, Mike Palovitch, twenty-four years at GSK, has a drug that is found in three medicines in the market, widely recognized as one of the leaders in kinda chemistry, but has a very let's lead with technology first chemist, and he's built a remarkable team.
• 24:17
  Vijay Shahani's been with us as VP of drug discovery. We've started to aggregate this really incredible group of individuals.And many of whom were early days and we've now trained.
• 24:28
  For example, Julie Owen, you know, now taking over a leadership role in drug discovery with Cyclica.
• 24:34
  On the computational side, on the development side, Andreas Windemuth, Steve McKinnon, people who've been with us since the very early days. So I don't need to make every decision. I don't lead through autocracy.
• 24:46
  I ask questions, I provide directions, I set the vision, but ultimately then on implementation, I have the best team in the world and a really incredible culture I think that we fostered to say, "You guys own it.
• 24:56
  You have the agency. Go do what you need to do." I love it. I love it. So we've talked about Cyclica's journey, a bit about your past and kind of where you sit in the market today.
• 25:06
  Let's turn towards the future and kind of what your, your next big moves will be and what you hope to have happen. Yeah. So what's the future look like?
• 25:15
  I come back to that fifty percent two thousand, seventy-five percent today, give or take, of drug discovery happens outside of the four walls of a pharma company. The time, the cost, it's just too much.
• 25:27
  And what continually strikes me, Logan, is the following important stat because this is what we are trying to solve for.
• 25:36
  Um, a biotech company that is working on one disease area, one program, has around a four percent chance of success of seeing that program actually materialize into market potential.
• 25:49
  Four percent-- A ninety-six percent chance of failure. A biotech company generally, traditionally was focused on one thing. They raise a bunch of money, they have the subject matter expertise, they go.
• 26:01
  Venture capital firms will invest in a couple dozen of them to diversify that ninety-six percent chance of failure risk. It doesn't have to be that way.
• 26:12
  We believe that on the back of our platform, we can create a portfolio of programs, each of which may have that one risk, but the entire company's not beholden to that risk.
• 26:25
  We can diversify that by making really smart bets across a broad range of assets, drug discovery programs that we control the destiny of. That's number one.
• 26:36
  Number two is in our industry, and this is the most glaring stat [sniffs]. There are thousands of diseases. Those diseases are driven by those proteins generally.
• 26:48
  Now, there are other drivers of disease I'm not g- I'm not talking about. It gets a bit unwieldy, but let's just talk about the protein space.
• 26:55
  Of the protein space, only seven and a half percent of proteins have enough data to be worked on by the physics-based technologies or the AI-based technologies.
• 27:06
  So ninety-two and a half percent of all proteins are generally untapped.
• 27:12
  So we said, "Instead of being another company focused on that seven and a half percent, let's go after the entire collection and go work on things that other technologies cannot, and where if we have that impact, there are patients that are just waiting by because it doesn't exist in the marketplace right now.
• 27:32
  So let's go after the undruggable space. Let's go explore the unexplored. Let's go drug the undrugged. And let's do that in creating a portfolio of assets." And we believe that that's the future.
• 27:45
  We are creating that biotech pipeline of the future. We're going to partner with pharma companies who are gonna develop those drugs.
• 27:51
  And by virtue of that, we believe we can have a huge demonstrable impact on human health. Uh, how has the industry's early response been to that massive vision? Oh, great question.
• 28:05
  Um, honest answer, a bunch of people call us fools, and that's why we're doing it, right? Like I- the IRR of drug discovery and development is near zero.
• 28:17
  It, it's like wh-when everybody says you're d- making the wrong decisions, this isn't gonna work, this is too risky, we're saying, "You're coming from the industry that's led us to this." Mm. Right?
• 28:29
  A sixty-time decline-- uh, sixty times decline in productivity over the past fifty to sixty years. Mm. An IRR close to zero. A four percent chance of success.
• 28:40
  And I have to listen to you saying that we're doing something that is the inverse of that. Of course, you're gonna say it doesn't make sense, and that gives us fuel. I will say like almost
• 28:53
  a hundred percent of people will say-- initially say, "That doesn't make sense to us." Now, of that hundred percent of people, about ten percent today are saying, "But I wanna learn more. That's really interesting."
• 29:08
  And I think that ten percent was one percent a couple years ago, ten percent today.
• 29:13
  I think it's gonna be thirty percent, forty percent over time, and I do believe that we will affect behavioral change in the way in which people think about the scientific opportunity and the commercial model.
• 29:23
  But it's not intuitive. We're trying to change something that's existed for sixty years. Yeah. And that's the battle that we, um, are-- you know, we have high conviction towards. Yeah.
• 29:33
  Well, knowing you, uh, having talked to you and, and, and watched you over the last few years, I have, uh, I have strong confidence in the, the research and the work that you're doing, uh, and the, the foundation that you're laying for this.
• 29:47
  And so I think you're, I think you're moving in the right direction. Um, I wanna close with a question that's maybe a bit more personal or more on the managerial side.
• 29:56
  You're one of the hardest working CEOs that I know, and yet I also know that you recently had a child. Mm.
• 30:03
  And it just got me thinking about just the general global mental health crisis within the workforce and our thoughts around, uh, creating, uh, mental space for employees.
• 30:14
  Anytime, uh, you know, a CEO goes through that themselves, I think it probably makes them think about, "Okay, how am I taking care of the mental health of my employees, giving them space for their family, et cetera, their life?"
• 30:25
  So, uh, I just wonder how, how do you stay balanced, uh, w- you know, in that new role as father and, and how has it made you think differently about your team?Bust out your cocktail. This isn't a short answer.
• 30:38
  Um, I'll say, you know, I went through mental health pro-- challenges back in twenty eleven, twelve, thirteen timeframe. Um, I went through my own challenges. Uh, I exhibited them in my own way.
• 30:52
  One of the reasons that I'm so convicted on or have such high conviction towards the mental health space and why we created Endogenex Biosciences is because that's a very untapped space.
• 31:03
  Now, I'm not gonna go down the therapeu-- I'm not gonna answer your question and nerd out on the therapeutic end. What I wanna talk about is the problem itself. Um, at Cyclica, we do kinda three things. One is
• 31:16
  coming out of COVID, and about six months after COVID, everybody was working from home, we implemented a work from anywhere policy. Um, now not e-every-- A bunch of companies have done this. Some have reverted back.
• 31:28
  For us, it is a core tenet of the organization where we trust our employees to be productive and to collaborate and to do things amongst their teams, and we as a company try and create the environment for them to have their togetherness and their collaboration.
• 31:41
  Are we perfect at it? No.
• 31:42
  But what we have seen is on a quarterly basis, we track our employee net promoter score, and as of Q1 of twenty twenty-two, our eNPS was-- or sorry, Q2 of twenty twenty-two, our eNPS was seventy-eight.
• 31:57
  Kind of best of the best. Like, people like it because they feel the flexibility, they feel the empowerment, they feel the agency.
• 32:03
  And as a result, we're-- we used to hear things about work overload or too much pressure or people's commute time and all these microstresses that would aggregate but would then be directed to the company have melted away, and I think satisfaction has gone up drastically, and mental health issues, we've heard, have diminished.
• 32:22
  The other way we've addressed mental health issues is we invest very heavily in our team.
• 32:26
  So we invested i-in a platform called Resilient Minds, and over the course of twenty twenty-one, we had monthly sessions and breakout teams where we'd have coaches and mentors come in and allow people a safe environment to talk amongst themselves about things that they were going through, and we had coaches walk people down a journey.
• 32:43
  Everybody was expected to attend. I attended, and it was a very open and honest environment to share things that you otherwise wouldn't share with your team.
• 32:51
  And the third thing is we have implemented unlimited wellness days at Cyclica. So every team member at Cyclica has, um, a minimum of four weeks vacation.
• 33:03
  We have two company-wide shutdowns a year, so a minimum of six weeks available to them, four discretionary, two f-forced. And then we do unlimited wellness days. Wellness days in our parlance used to be called sick days.
• 33:16
  Nobody ever wants to be called sick. Nobody wants to call and say, "Hey, I'm sick. I'm not..."
• 33:22
  But if you're saying, "I want to be better," just changing the language, we started to see people were taking time off and coming back rejuvenated.
• 33:31
  And then coming to your question on parenting, I didn't appreciate what it was like to actually do something meaningful and give everything to it while also having something on the side, which is a family.
• 33:41
  My wife Sabrina has been with me for eleven years. She knew Cyclica was the thing, and she was there, and we're great relationship.
• 33:48
  But as soon as Noah was born, it was like my entire philosophy was like the two of them was the most important thing. Cyclica, if I focus on that, I would be the best person at work. It's made me a bit of a softy.
• 33:59
  My-- She has me wrapped-- Noah has me wrapped around her little pinky, and people at work have said I've just like chilled out and become a little bit more easy to deal with on a day-to-day basis.
• 34:09
  I think it's been the best blessing, um, but also made me deeply empathetic towards what it's like to be a parent, whether you're a father or mother in the workplace and trying to do stuff, and I didn't have that empathy before.
• 34:20
  Hmm. I love it. I love it.
• 34:22
  Naheed, um, I've always r-- uh, appreciated your vulnerability and your, your clarity of vision, and so I really appreciate you sharing with me today, uh, and I know our listeners will get a lot out of it as well.
• 34:36
  Logan, thank you so much. I appreciate you and Startup Health and obviously everything that you do to evangelize the healthcare community globally.
• 34:43
  We'll be watching Cyclica and seeing what you build next, and we'll be excited to hear, uh, your next report. Thank you so much. All right, be well. Bye. Take care.
• 34:52
  [outro music] Thanks for listening to Startup Health NOW. Startup Health invest in health transformers around the world who are dedicated to achieving audacious health moonshots.
• 35:03
  If you wanna learn how you can join this community of entrepreneurs, or if you wanna connect with one of our four hundred companies, go to startuphealth.com.
• 35:11
  If you'd like to learn how you can invest in our Health Moonshot Impact Fund, go to healthmoonshots.com. Thanks for listening to Startup Health NOW. We'll be back again with another episode next week. [outro music]