Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Ep. 15 Jonathan Steckbeck on What Biotech Can Learn from Poker
51 Plays3 months ago
Jonathan Steckbeck is CEO of Peptilogics, where he is charting a path to therapeutic and commercial success in antimicrobials. Listen in as he describes how AI is impacting their work, and how he draws inspiration from games like poker.
Recommended
Transcript
Introduction to Mighty Spark Communications and the Role of Storytelling
00:00:00
Speaker
Hi, everyone, and welcome to Spark Time. I'm Dani Stoltzfus. And I'm Will Riddle. Of Mighty Spark Communications. Our mission is to use scientific innovation to drive transformative change. We believe that compelling storytelling is the most effective tool we have in our arsenal to motivate and inspire audiences to invest themselves in audacious goals. We are scientists by training, storytellers by experience, and entrepreneurs by nature. Let's get started.
Conversation with Jonathan Steckbeck: Antimicrobial Resistance and Business Challenges
00:00:28
Speaker
So our conversation today with Jonathan was really memorable. You can tell he thinks really deeply about how he and his company communicate, as well as the practices that they bring into the company. He's really mindful about the tools and ideas that they incorporate. And on top of that, he's such a lovely person.
00:00:46
Speaker
completely agree. One point I was really interested in during this conversation was the discussion around antimicrobial resistance, how Jonathan and peptologics are charting a clear path in this territory that in other cases can be difficult to find willing investors because of reimbursement structures typically. But also, he thinks that they can address even bigger problems in that space. So that was really neat to hear.
00:01:13
Speaker
the end of the conversation, he really reflected on how difficult it is to build an innovative business and his thoughts on others in the same situation has really been useful for him. I like him sharing his thoughts about that because it's so helpful for early stage founders or anyone building a business in this space.
00:01:31
Speaker
Yeah, I love that as well. So let's get into it.
Peptologix's Mission and Jonathan's Career Path
00:01:34
Speaker
Today we're joined by Jonathan Steckbeck. He is the founder and CEO of Peptologix, a clinical stage biotech platform company pioneering the development of best in class novel peptide therapeutics. Under his leadership, Peptologix has developed a computational drug discovery platform to speed the design and development of novel peptide drugs across a range of therapeutic targets.
00:01:58
Speaker
The company is now rapidly advancing a pipeline of anti-infective therapies in focused indications and expanding the application of its platform to new disease areas. so Jonathan has a PhD in biochemistry and molecular genetics, as well as an MBA from the University of Pittsburgh.
00:02:15
Speaker
So Jonathan, let's start out talking about your path to biotech. because like It seems a little unusual to me because most people suffer through their life science PhD, realize they made a mistake, and then go back and do their MBA. But you did it the other way around, and I'm just curious as to why you did it that way.
00:02:33
Speaker
Yeah. So it is and it is an unusual path i've I've learned. And so really the way that I arrived at this, maybe it's useful to just provide a little bit of background. And so yeah I had done my undergraduate degree in microbiology. And during that time, I had interned at a biotech company ah for a few summers.
00:03:01
Speaker
and you know got to do a bunch of lab work there. But one of the things that I noticed was that while the scientists got to do all the interesting work, they didn't ultimately make the decision on what work was being done. And that was really the first and the first indication that you know just a scientific background, if you wanted to go into something like biotech or pharma,
00:03:27
Speaker
might not be enough if you wanted to have an impact at the highest level. And so when I graduated college, I worked for a few years and then I had an opportunity ah to go back and get an MBA because you know informed by that time, I knew that was something I wanted to be able to do. ah So got the MBA and then went and got another couple years of work experience.
The Intersection of Science and Business in Biotech
00:03:52
Speaker
And then you know the kind of the other thing that and makes sense in a science-driven field like biotech or biopharma is having a deep scientific background is is pretty important. and so I took an IRA, maybe I entered the PhD program with a little bit of a unique perspective relative to I think most people who enter. and It was really with the idea of being able to or having the opportunity to do deep technical diligence on a project, you know hopefully selecting the right project, but finding something that i could yeah it looked like it would be commercializable
00:04:35
Speaker
have the time to do the deep technical diligence on it and then you know if everything worked out well and and hopefully be able to guide that program or project through to something that looked more you know like a pharma type product, i.e. a drug, ah to be able to then get into the position where it made sense to to take that out.
Investment Challenges in Antimicrobial Projects
00:05:00
Speaker
Yeah, really interesting. I mean, I would say my experience was kind of a little bit different. It took me many years of working to realize that I wasn't in a decision making capability doing what I was doing and I needed to fix that. So I like that we share that goal of really being able to drive the direction and the strategy of what happens with the sites. I think you and I share that in common.
00:05:23
Speaker
Yeah, well, I don't know how you feel about, you know, I entered the PhD program knowing I hated lab work. yeah It was maybe doubly. Yeah, it was maybe doubly, ah kind of a double burden there.
00:05:38
Speaker
I want to move to peptologics for a moment, Jonathan. So, peptologics has been successfully charting a difficult path. And I say that knowing that, you know, as a microbiologist, antimicrobials, you know, are close to my heart, but we've we found and we've heard that there are tough space to find investment. So, would you talk to us about the the secret tools that you use to get investment in this space?
Peptologix's Focus on Unique Diseases and Solutions
00:06:05
Speaker
And I appreciate you recognizing you know kind of how difficult the space is. it say I'm not sure there's any secret tool. It was really, I think fundamentally driven by our approach kind of as a company and then how that is kind of instantiated in our program selection. And so while yes, um our product is an antimicrobial,
00:06:33
Speaker
our indication is really think i think drl it's really unique relative to and but what do we typically think of as antibiotics. right and so What you're referring to is the difficulty and and and a real difficulty that exists in gathering investment for, say, standard antibiotics, things that you would you know likely recognize if you get an ear infection or a- Low price, low volume assets. Exactly.
00:07:08
Speaker
exactly and so while we hear you know all and there's and There's a lot of noise and and it's real around the idea of antimicrobial resistance. It just hasn't manifested in such a way yet that the typical development paths
00:07:25
Speaker
allow you to capture essentially the data to be able to drive usage and more importantly, reimbursement, right? So the commercial prospects for, again, what I just try to think of as kind of typical antibiotics are are not good. And then that drives the difficulty in in gathering investment. So what we are you know what we've done is really found a unique a unique condition, right, a unique indication, a unique disease where all of the challenges that are typically associated with antibiotic development don't exist. And so we're developing, so our lead program is a drug, it used to be called PLG0206, now it's called xaloganin, for treating what are called periprosthetic joint infections.
00:08:16
Speaker
and or more broadly, uh, medical device infections, starting with periprosthetic joint infections. And so what those are is imagine if you get a knee replacement, right? You know, your, you or your parents get a knee replacement. Most of the time that's going to go really well, but sometimes they get infected and when they get infected,
00:08:42
Speaker
It's a catastrophic problem because what happens is is the typical antibiotics that we take for those other types of infections don't work. and They haven't worked for the last 40 years. right and That's a function of the different way in which the bacteria grow on the surface of the of the device.
00:09:02
Speaker
and so you know We can dig into this a little bit more, but because there is no drug that works, right the standard treatment course here is multiple major surgeries, we have an opportunity to bring the first therapeutic to market and by doing so, address a lot of the challenges that are associated with typical antibiotic development. so Maybe with that overview, we can dig in wherever you see fit.
00:09:30
Speaker
Right. So I'm curious, my and forgive me if this is a naive question, but but do you envision this as an entry point to um a broader antimicrobial market? So we do. And so we think about it. So maybe just kind of lay out the way we're thinking about our longer term strategy is really around
00:09:56
Speaker
first treating, again, like I said, device-related infections, because whether it's a joint or you know kind of a hardware that's you know used to stabilize spine, like fixation associated with trauma, implanted catheters, et cetera, all of those suffer from infections. And those infections, because of the way in which the infection occurs,
00:10:24
Speaker
all of those are, again, not well treated or not treatable with standard antibiotics, and so they're all surgical problems. So we see that as the first big opportunity because, again, the economics are such that oh you have actually a there's There's a huge benefit to finding something that can reduce surgeries. right Surgeries end up being the most expensive things that happen in hospitals. and so If you can reduce the number of those that are required to and of bring the patient to good health again, then you actually have an incentive for and of all stakeholders to use them. right They're better for patients, they're better for surgeons, they're better for insurers, and they're better for the hospitals.
00:11:10
Speaker
And then you know that allows us to create a market in a place where there hasn't been one writing and capture those economics. But then over time, what we'd really like to be able to do, and some of this is at you know this ends up being a little bit out of our hands, is get ourselves in the position or be in the position where we can start to treat some of those other, what we think of as more typical, typical kind of bacterial type infections. The places we know we're going to need new drugs in five, 10, 20 years, but the incentives just aren't there to develop them right now.
AI in Drug Development: Potential and Limitations
00:11:51
Speaker
Speaking of tough spaces, and I know we just we had a really good explanation from you, Jonathan, about the antimicrobial space, but I want to switch topics and talk about AI now because my understanding is that you know your Nautilus platform is based upon AI algorithms and aids in your like drug discovery process. and We've heard a lot of commentary recently around the fact that AI and drug development development excuse me isn't living up to expectations. I'd love for you to share your thoughts on the current status of AI and drug development. and Do you see at any point it replacing people in the process? and I'd love to hear your ah thoughts on that.
00:12:35
Speaker
Yeah, yeah, absolutely. And so maybe I'll start with that last one, because I think that's one of the things that we want to be really careful with, both kind of ourselves, but I think as a broader industry is, you know, we don't ever, at least not for the near, you know, in in the very near term, see AI as a replacement for humans. We think about it as a way to augment human abilities, right? And, you know,
00:12:59
Speaker
We all spend a lot of time in training formally and then informally on the job to become really good at drug development. And there's a lot of heuristics that are built up that are actually really valuable. And so we see AI really as a and force multiplier to be able to allow scientists to be more creative in their jobs right and to be able to explore a larger space more effectively than really act as a replacement. I think that's the that's the first thing and then maybe coming back to
00:13:36
Speaker
you know how do we think about AI and how that fits into and of maybe the broader zeitgeist for it? right and so I would agree. and i think you know what What we end up seeing often and maybe what we're seeing right now is kind of a playing out of a hype cycle. right and so you know I would say maybe say in the last 10 years or so when we had what I'll call the first wave of AI focused companies in pharma. And there was, I think two things occurred, right? One was building off of, well, maybe the more technology focused kind of AI, right? Where
00:14:26
Speaker
the general thought process had been more data, more data, more data. If I can collect enough data, then now I have a suite suite of algorithms or a toolbox by which I can manipulate that data to get answers. You're going to get answers out of it. right And that's you know most recently and of evidenced by OpenAI and some of the other other um large language models where I just collect a ton of tokens, right which is essentially a lot of data.
00:15:02
Speaker
And if I can throw more compute power at it, well, I can now figure out the structure of language and I can figure out all this information. And now when you ask it questions, you get a lot of good information back. You still have to be careful with it, but you get to that of good information back. And I think that's where a lot of companies started and saw parallels.
00:15:21
Speaker
which was just, I'm going to build a data generation machine, I'm going to collect as much data as I can possibly collect, and then I'm going to throw algorithms at it, and we're going to solve and answer biology. And I think what we're seeing now is a reflection of that being and maybe the wrong way to think about it. And so what i mean what I mean by that is we came at this from a slightly different angle. right And where we started was, okay, let's let's think about what biology is. right Because what we're really trying to do with the use of AI is
00:16:04
Speaker
make the process more efficient. and Because we all understand and you can look, you know do a quick Google search and figure out that the cost the fully capitalized cost of any new drug is somewhere between two and $6 billion dollars depending on you know depending on which ah which study you look at. So it's an enormous cost. And the time is somewhere between 10 and 20 years, right just kind of as a range.
00:16:30
Speaker
a lot of that failure occurs early on. And so what we'd like to be able to do is shrink both the time and cost for drug discovery and development. Now, what that means though is that you're, or maybe not what it means, but the realm in which we're playing is biology, right? And biology is incredibly complex. And so and as we approached it, it was really from a If we think about the complexity of biology and just start to try to draw some numbers, you very quickly get to numbers that are that humans don't really understand anymore. right like It's not out of the realm of possibility to think about the number of unique interactions or something like that being in the order of 10 to the 200th.
00:17:22
Speaker
right and so that It's an enormous number that nobody really has ah an intuitive grasp on. But what it means is when you start to think about it in the context of drug discovery, it's okay. So say you have a library and say you're generating a really large library and you're doing something like phage display. right and that So that and of limits the number of the types of programs that we're talking about, but it's a good way to
00:17:48
Speaker
provide a kind of a framework to think through this. And so say you do a library with a trillion molecules, right? So that's what, 10 to the 12th, right? Or you do 10 to the 14th, whatever, it doesn't matter what the number is, but say you do 10 to the 14th or even 10 to the 15th. So you've got 10 to the 15th unique molecules, but you're working in a space that is 50, 60 orders of magnitude larger. So you're effectively collecting zero data. right What you're describing is really 0% of what actually exists. And so we believe or or we took the approach that
00:18:38
Speaker
a brute data force a brute force data collection was never going to get us to the place where we could adequately describe the full complexity of biology. And so we started from the ground up of basically saying, if biology is this complex, where are the areas where we can try to be smart about the algorithms that we use or apply it to places where maybe biology has done some of the work for us to try to understand and and pull out commonalities or underlying principles that essentially guide all of that complexity.
00:19:26
Speaker
It's a complicated topic, so you know I'm happy to dig in anywhere to make that a little bit more clear. so it's It's so interesting how how peptologics is being so thoughtful to to use it as a tool to do things that that you know people can't do rather than you know accelerating the things that that people can do. I think you're absolutely right. that it It's a tool that that's going to accelerate the process and um we're excited to see where Peptologix takes us. But I want to continue on this theme of AI. I think this is such a great example of cross-pollination um between different fields, between especially right now between tech and biotech.
00:20:09
Speaker
um But I'm wondering if if there are any other examples, Jonathan, of ideas or practices from a different industry that have inspired breakthroughs or improvements in your company in any other way.
00:20:24
Speaker
Yeah, funny need enough, it actually came from games. so like video and ah po actually poker yeah and so and so this was It really informed, again, kind of how we not only how we think about doing things, but also whether or not it's actually possible.
00:20:45
Speaker
right And so what I mean by that, or maybe to put foot put some specifics to it is, so I think it was the late 90s or maybe early 2000s, AI but beat humans at chess. And so if you think about what chess is, it's a perfect information game with a ah really well-defined space. and And so you can essentially brute force that.
00:21:10
Speaker
right And then likewise, on orders of magnitude larger, I think sometime in the mid 2010s, computers started to beat humans at Go, another fixed space, but perfect information game. You can know all of the possible moves. But one of the folks on our team in the late 2010s worked on a team at CMU, or as part of a team at CMU, that ended up
00:21:42
Speaker
creating the first AI system that could beat humans at poker. And so the reason that's important is twofold. One was the fact that poker is an incomplete information game. ah You know your cards and you know a few cards that you can see.
00:21:58
Speaker
out there, but you don't know your opponent's cards. right So it's an incomplete information game. And there's a little lot of... So incomplete information, it's high complexity. right There's a lot of potential lot of potential and options or outcomes, if you will, at any one given or at any time given what's on the table. But the complicating factor is that your opponents can lie to you. right And they can lie to you in
00:22:30
Speaker
strategic ways, right? So there's there's a strategic ways, non-strategic ways, right? So there's just an added element of complexity that has nothing to do with what's actually on the table. It has to do with human behavior. And when I saw that computers could essentially capture the inherent uncertainty built into the game but then compounded by human behavior,
00:23:00
Speaker
it became clearer that if we took the right approach, we could probably solve something as, or or get closer to solving something as complex as biology using the right types of approaches.
00:23:12
Speaker
yeah That's very cool. I think I just watched a Star Trek episode where the Android beats the humans at the computer, so. but fact It could work well. I'm going to like completely change topics down on you, Jonathan, and talk about trust.
Communication and Trust in Biotech
00:23:31
Speaker
And the reason I want to go there is because Will and I talk a lot about trust and especially trust in terms of communication. And we really strongly believe the importance of trust building trust with your audience. And that could be investors or pharma partners or even patients in the in your case. And so when you think about good communication, how important is trust to you and what other core beliefs do you hold around good communication?
00:24:03
Speaker
Yeah. i mean Trust is is hugely important as you recognize and it sounds like you've built kind of your thesis and ethos on. I think if you you know particularly in this type of business where at the end of the day, we're judged on you know kind of call that more objective outcomes, but those objective outcomes, there's a lot of subjectivity around them, right? So clarity and transparency in the you know first and foremost, the data and around how you're arriving at it is is crucial, right? Because I think yeah
00:24:55
Speaker
you know and coming maybe from, or maybe not growing up inside like large farmer, right? It's interesting to see how, and I almost hate to say this kind of out loud, but I will. and that you know it you can
00:25:14
Speaker
There's so much variability within the range of outcomes that programs, if you're not honest about them and you're not honest with yourselves, you could push a program further than you probably should. And it falls within the realm of what's reasonable.
00:25:36
Speaker
And so you know being very, like I said, clear and honest with outside you know outside parties, but also with yourself around you know what what a success criteria needed for any given program are critical because it informs both the time that you're going to spend, but then you know your potential investor's money, your partner's money, everybody's time that's along that program, the hope that you give to patients.
00:26:06
Speaker
so i think you know if you if you It's such an interconnected industry with so many different unique stakeholders that don't always have the same incentives. If it doesn't center on the trust around what it is that we're building and what we think the ultimate outcome can be, it's so easy to get lost in in that web of different incentives.
00:26:36
Speaker
Yeah, that is such a thoughtful response, Jonathan. Thank you. So
00:26:43
Speaker
Right now, peptologics is you know communicating that you have ah a great solution for a field that struggles to sometimes fight gain investment. and You've described the complexities of the product that you're working on that that make that possible. and you know You shared the the goal that that you share with investors and and with everyone on the team of of creating a great therapeutic for for this area. so You also have the added complexity of explaining AI in the process. and so you know Clearly, it's all coming through like you're a great communicator. so Tell us, what what is your strategy around making complex information accessible to the non-experts in the room?
00:27:29
Speaker
I appreciate you yeah thinking that it's ah that I'm a good communicator. it's true I think the what we try to do is i think it's it's one of those situations, right and I think it was maybe, I can't remember if it's attributed to Feynman or Einstein.
00:27:53
Speaker
but the idea that you know if you don't really understand something, you can't possibly explain it to someone else. is is is the way that we try to approach it. And so what we do is, and and and I'm not sure we're unique in this regard, right where it's understand everything down to the base level, whether it's on the clinical program, you know talking to patients and understanding their
00:28:24
Speaker
the challenges that they face given the you know in the current paradigm and what they've had to experience over you know through their infection or you know whatever that is, talking to surgeons, talking to hospitals, talking to payers, just trying to really understand everybody's challenges, motivations, and then find the thread that connects all of those things.
00:28:53
Speaker
and then you know and then doubly trying to find the simplest way to explain to each group, but all groups, why a solution that can address these unique points for all of the stakeholders is actually something that is not only useful, but that it will you know create economic benefit for the entire system. is So it's what we try to do. And I guess the the way we do that is often through,
00:29:31
Speaker
ideally we try to find metaphors, right? Because I think one of the challenges that we have both in, you know say, PJI, i which is you know our first actual indication,
00:29:44
Speaker
and then within within AI is that they are
00:29:50
Speaker
sometimes they're and They're incredibly complex, but there are also sometimes things that have either different preconceived notions in the case of AI or sometimes nobody's ever heard of it, right like in the case of PJI. And so explaining how it's similar to something maybe they understand, but then also equally sometimes how it's different than something they do understand and maybe they don't like.
00:30:18
Speaker
is has been, I think, reasonably effective.
00:30:25
Speaker
Yeah, it makes a lot of sense. I always love a good analogy when I am trying to explain something to a non-expert audience and it sounds like, you know, your metaphors do the same thing. So i I can appreciate the value of that. I think there's a lot of power in being able to find the words that really connect you with your audience and speaking to them in the language that's familiar to them so that they can understand, you know, what the importance and the gravity of what you're trying to communicate. So you know, I feel really aligned with you on that, Jonathan. And I guess changing topics again and thinking towards the rest of 2024 and upcoming into 2025, I'd love to hear what you're excited for and what's happening over at Peptologics and any other exciting news you'd like to share with us.
Progress in Peptologix's Registration Trial
00:31:16
Speaker
Yeah, I mean, so for us, I think the one of the biggest things, and the thing that's driving a lot of ah the company right now is moving forward with our, our registration trial. And so earlier this year, we finished up a ah phase 1B trial where we treated patients with our drug. So we treated PGI patients and we got a really great outcome there.
00:31:43
Speaker
just you know numerically, we had ah you know like an 86% improvement over standard of care in outcomes for these patients. And so yeah, I mean, it it was really exciting because again, there's really no, right now patients are going through multiple surgeries, failure rates somewhere around 50%. And so being able to essentially give a person their life back by not having, you know essentially by not having to go in for two or three surgeries, which is, you know this is really a life-changing event for people. So being able to give them that mobility back, being able to give them their life back is is incredibly gratifying. And so now with those good results, you know we are
00:32:29
Speaker
close to wrapping up a funding round to be able to take this the drug through the registration trial and then hopefully be able to bring it to to patients here within the next few years. That's great news. Very exciting.
Advice for Early-Stage Biotech Founders
00:32:45
Speaker
So, Jonathan, this has been a really enlightening conversation, and we really like to cap the discussion by asking, if you were talking to the founding team of an early-stage biotech, what is the single piece of advice that you would give to that team?
00:33:05
Speaker
Persevere. I can do it in one word. Great. I think it's... you know you It's one of those things where, you know I guess maybe speaking from personal experience, you go in knowing it's going to be hard. right You can objectively, and I think maybe one of the benefits we have as trained scientists is typically we try to evaluate all the data that you can find right to inform your decision. So I went into this knowing that yes, drug discovery and drug development is really hard.
00:33:41
Speaker
But I don't think you can possibly, and I don't think it's it's this way anywhere, right? I don't think it's possible to understand from a descriptive set of accounts about how challenging things can be or can become. And so just, I think, knowing that ah Anything you're going put through is probably normal, right or has at least been that or has least happened to someone else before. is I think that first that first little spark of, hey, I can get through this because other people have done it. There's lots of little things that can happen that
00:34:29
Speaker
no
00:34:31
Speaker
can cause you to to kind of question, you know is this the right thing to be doing? And so just being able to kind of persevere through those and challenges that are inevitably going to arise is, I think, the one the one piece of advice, I guess, I would give to readers, Shackleton's book.
00:34:51
Speaker
Yes. Yes, absolutely. I think that's fantastic advice and and Jonathan, it's it's really been a pleasure having you on the podcast. So, I want to say thanks very much and and we're excitedly following along with Peptologics. I can't say it enough. It's really a cool strategy to develop antibiotics around specific use cases that are viable markets like PJI in the case of Peptologics.
00:35:14
Speaker
and then have the ability to translate those learnings to larger markets as the need arises. What a great innovation and strategy. Oh my God, Will, you are such a nerd. like Seriously. But whilst we're talking about AI approaches more and more on the podcast, I can appreciate why you're getting so nerdy. And I really liked hearing Jonathan's take on what it means to peptologics to use their tool to let the scientists be more creative and solve even bigger problems than before. For me, that sounds like the best use of any technology.
00:35:50
Speaker
so Jonathan made excellent points around building trust, both within the company and with external parties, in that there is such a range in outcomes that are reasonable within biotechnology. Additionally, there are so many stakeholders with different incentives, and it's really imperative to be clear about the ultimate outcome the company is building so that all the parties are aligned and pushing for the same goals despite having those different incentives.
00:36:15
Speaker
Yeah, i I always love a good conversation about trust. You know that well. And speaking of trust, I'm going to switch topics and not talk about it and talk about his philosophy on communication because to understand everything at its base level, as well as its motivation and its challenges, I really like that concept because we know from experience that this is much easier said than done. And we're always saying to our clients and anyone who will listen really that you really need to know your audience when you're speaking to them.
00:36:45
Speaker
I think Jonathan's getting at the same point. It's that you need to know your audience and explain in the simplest way how your solution can address each point. Well, what a great conversation we have with Jonathan. Join us next time as we continue to power scientific innovation with storytelling to drive transformative change and solve our most demanding challenges.