A Rescue Model For Shelved Drug Candidates With Annette Bakker, Ph.D. And Andrew Lo, Ph.D.

On this week's episode of the Business of Biotech, Annette Bakker, Ph.D., CEO at the Children's Tumor Foundation, and Andrew Lo, Ph.D., Finance Professor at MIT's Sloan School of Management and Director of the Laboratory for Financial Engineering, talk about the methods and economics for turning abandoned drug candidates into new therapies for patients. Annette and Andrew share the story of Gomekli, a formerly shelved Pfizer drug candidate spun out into SpringWorks Therapeutics, and approved in 2025 for NF Type 1 tumors (Merck KGaA subsequently acquired SpringWorks for $3.9 billion). They also detail plans for bringing more shelved candidates back into development, using a shared data infrastructure and a matchmaking tool to pair interested drug developers with investors and funding. 

Access this and hundreds of episodes of the Business of Biotech videocast under the Business of Biotech tab at lifescienceleader.com.  

Subscribe to our monthly Business of Biotech newsletter.

Get in touch with guest and topic suggestions: ben.comer@lifescienceleader.com

Find Ben Comer on LinkedIn: https://www.linkedin.com/in/bencomer/

Transcript

Welcome And Repurposing Vision

Ben Comer 0:03

Welcome back to the Business of Biotech. I'm your host, Ben Comer, Chief Editor at Life Science Leader, and today I'm speaking with Andrew Lo, Ph.D., Professor of Finance at MIT's Sloan School of Management and Director of the Laboratory for Financial Engineering, and Annette Bakker, Ph.D., CEO at the Children's Tumor Foundation, a nonprofit organization focused on accelerating the availability of treatments for neurofibromatosis and schwannomatosis or NF, a group of rare genetic conditions that cause tumors to grow on nerves throughout the body. I'm pleased to have Annette and Andrew on the show today to talk about a new repurposing model for identifying and developing NF treatments, a model that has already notched a big win with the approval of Gomekli for NF Type 1, a formerly shelved Pfizer drug candidate spun out into SpringWorks Therapeutics, which was in turn acquired by Merck KGaA for $3.9 billion in July of 2025. Andrew and Annette hope to replicate this success by building a marketplace for discontinued drug assets and a shared data infrastructure and a structured matchmaking mechanism to connect to connect asset holders, developers, and long-term capital funding. Andrew and Annette, thank you so much for being here.

Andrew Lo, Ph.D. 1:27

Thank you. Thanks for having us.

Ben Comer 1:29

Absolutely. It's an interesting idea, and I want to get into it. But before we do, uh I have a couple

Finance Meets Drug Development Risk

Ben Comer 1:35

of questions just on background. Andrew, um, you're a finance professor at MIT. I'm interested in what caught your interest initially about repurposing discontinued drug assets.

Andrew Lo, Ph.D. 1:49

Well, that's a great question. And I ask myself that all the time. How did I get involved? Because I'm really deep into it now. Um it started about 15 or 20 years ago when a number of friends and family were dealing with various kinds of cancer. And through that process, I realized that finance actually plays a pretty big role in drug development. Uh, in some cases, the wrong kind of financing or not enough financing preventing good drugs from ever reaching patients. So I just started learning more about the drug development process. And the more I studied, the more I realized that, you know, we have a lot of tools in finance that can actually make that process more efficient. And so since then, I've been working with experts in the field like Annette, trying to understand how to apply these concepts of financial engineering, things like portfolio theory, diversification, risk management, and so on, to situations where we don't have enough drugs for patients in need. And drug repurposing is one of the very most important areas for finding these assets that are diamonds in the rough.

Ben Comer 2:54

Was there a, you know, like a specific financial engineering problem that that you recognized and kind of set out to solve?

Andrew Lo, Ph.D. 3:03

Well, sure. The main problem with drug development is risk. We don't know whether or not a drug is actually going to work. So we have to do a lot of human testing. And that process is grueling, very, very expensive, takes a long time, and you know, in some cases, there's no way to estimate what the likelihood of success is. And so all of those conditions, to me, spell the perfect scenario for using tools that reduce the risk and allow for multiple shots on gold or uh uh the idea of portfolio theory. That that's where I think finance can play a very positive role.

Ben Comer 3:40

Uh just for listeners who maybe aren't aren't familiar, can you give me a kind of highline summary of what portfolio theory means?

Andrew Lo, Ph.D. 3:49

Very simple. Portfolio theory is the old adage of don't put all your eggs in one basket. What you want to be able to do is to spread your bets so that you have these multiple shots on gold, use a hockey or a soccer term. And as a result, you're more likely to have one or two winners in a large enough portfolio. And given the incredible success that drugs offer to investors when they get approved, you only need one or two successes to pay for all of the different trials. So not putting all your eggs in one basket is basically the idea. And, you know, not everybody agrees with that, of course, because you know, Warren Buffett has famously said he puts all his eggs in one basket and then he watches that basket very carefully. And, you know, if you're Warren Buffett, I I think more power to you. You can do that. Most of us are not Warren Buffett, and in the case of drug development, it is almost impossible to tell at the very early stages of drug discovery whether or not you've got a winner or a loser. And so in those circumstances, portfolio theory is really ideally suited for funding that kind of investment.

Ben Comer 4:56

Yeah, and we're gonna get into this some more, but it portfolios uh kind of counterintuitively to what you're saying sometimes shift companies away from promising assets because they don't match up very specifically with uh with what they're trying to do or how their sales forces are organized, a number of different reasons. We're gonna talk about that in a few minutes. Um but Annette, what what circumstances

Building The NF Discovery Engine

Ben Comer 5:20

led you to NF specifically in the in the Children's Tumor Foundation?

Annette Bakker, Ph.D. 5:25

So, first of all, thank you for the question. I was a pharmacologist, actually, in the industry. So I worked for both large pharma and then in a biotic. And one of the things that I discovered was that the two ecosystems that you would really need to work together to get drugs and treatments to patients were pretty siloed and actually not really working together. The academic institutions have wonderful cell and animal models. The companies have wonderful drugs, but these two ecosystems just didn't collaborate. And it's not because people are bad, it's just because the incentive system is opposite. In the industry, you need to keep stuff secret as long as possible. In the academic institution, you need to publish as quickly as possible. So there is this whole, there is a little bit of a disrespect and a mutual shooting at each other that is happening. And I was like, instead of accepting that, this is one of my hobbies, is to say, okay, there is this axioma that everybody just accepts, even if it's not working. And what if we question that axioma? And so that is what made me join the Children's Tumor Foundation and really creating partnerships between academia, industry, regulators, clinicians, patients. So really seeing, like, okay, only together will we be able to really solve the problem. And it is fascinating to see how incredibly powerful it just is if you can work together on a problem that is affecting people and put the human face on the problem. It's just incredible what people are willing to do. So I'm still convinced that in healthcare, 90% of the people in healthcare are in healthcare for the right reason. So that was my starting point.

Ben Comer 7:12

And the the Children's Tumor Foundation describes itself on the website as an NF discovery engine. Is that what you're talking about? This collaboration with biopharma companies, researchers, and others?

Annette Bakker, Ph.D. 7:26

So yeah. So what we, in fact, what we have really been thinking about as the foundation is instead of saying, okay, we're giving out grant money to all good ideas, which I think is absolutely essential. However, there is also another aspect is that nonprofits can do things that nobody else can do. We can meet with pharma, with regulators, with clinicians, with the patients, back with pharma. We have that freedom to operate, if you want, um, to talk to everybody and anybody and convene the right people to make things happen. That's the first, let's say, value proposition that we have as a foundation. The second value proposition is that we really know everybody. And that was one of the big successes of the Go Meckley story is that being in the foundation, I knew two things. I knew that Pfizer was going to shelf that asset on the one hand, and I knew that our clinicians had positive clinical data on that very specific asset. But these two didn't know that. So then it was, in fact, the whole clue was how do we bring these people together to talk to each other and actually know that we have great clinical data, that it would be a mistake to shelve that asset and that there was a real opportunity here to bring that asset to the clinic. So the other thing that we have done in that same philosophy is we have been thinking about, okay, we don't accept the axiona that it takes 10 to 15 years to develop a drug and whatever, how many billion dollars and how many whatever failure rate. We were like, okay, what if we crush the time and we build a preclinical hub that if a drug comes, we can immediately test it in the models. We have pre-negotiated agreements with either CROs or academic institutions. And then we have built platform basket trials, which are innovative clinical trials where instead of looking at one manifestation for each drug, you can actually randomize your drug to three, four, five different manifestations at the same time. So you just increase your chance to win um a drug and win an approval. So what that is what our engine is, is really the pre-clinical hub coupled to the um platform basket trials.

Ben Comer 9:38

Interesting. Oh, and and human

Gomekli Case Study And SpringWorks

Ben Comer 9:40

relations, human connection, kind of uh fitting it all together, making uh making those important connections uh is a key piece of this. How Andrew and Annette, how did you meet? Uh how did you start working together?

Annette Bakker, Ph.D. 9:53

I had a I had a crazy idea. So after after the Go Make Lee approval and the 34X return on investment of Spring Works by selling to uh the Merck KGA, I was like, we have to think how we can do more of this. So I was invited by the World Orphan Drug Conference to organize a panel, and I kind of took my courage in my hands, and I was like, if there is a few people that can really make this happen, it is Andrew, it is Tanisha Carino, who is a uh policymaker, and it is Frida Louis Hall, who was the former chief medical officer. So I invited them to a panel, and on that very specific panel, Andrew committed. He was like, let's do it. So we walked off the panel, and that was that was it. So and it has been a wonderful collaboration.

Ben Comer 10:47

Well, uh Gomakli is obviously uh an important case study uh supporting how this model can you know be successful. Um, Annette, could you maybe tell me how you identified uh what I guess at the time was called Merdatimit or MEDEDANIB in you know Pfizer's shelved asset catacombs, you know, how how you how you found it, how you licensed it for development.

Annette Bakker, Ph.D. 11:13

We didn't license it. So this is in fact, no, we did not. So we um well we were on the site. We were the um research foundation that helped make it happen. But at that time, we didn't make an investment. We just wanted to make this happen. We are in this game to get treatments to patients, right? If we make money on the way, great, but that's not our objective. So that also made us as a very neutral convener and a very neutral partner in this entire exercise. So I met, in fact, with Lara Sullivan, who was at that time working for Frida Lewis Hall. Um, I met her at an event and we started talking, and she said, you know, we have maybe some shelved assets that or some assets we're thinking to shelve. And then I invited her to the conference, and she was like, Oh my God, I feel like I'm sitting here in a Pfizer portfolio meeting. I see all these drugs passing by. And then we started talking. And in fact, over a couple of meetings, I figured on the one hand that Murda Methanep was being or going to be shelved by Pfizer because there was a little bit of a competition. Well, a little bit like in every you know for-profit entity, there was a competition between AstraZeneca with Golme, uh Coselico and then Myrda Methaneip from Pfizer. It was sorry, it was clear that AstraZeneca was going to win the race if you want to the approval. Um, so that was done as a consequence. There were quite some companies that had make inhibitors that kind of said, okay, we're we're giving up. But the chemo type of Myrda Methaneip was so different from Coseligo that I insisted that this was really a good opportunity for our patients. And so here we are.

Ben Comer 12:55

What was the AstraZeneca? What was the lead indication for the AstraZeneca product?

Annette Bakker, Ph.D. 12:59

NF1.

Ben Comer 13:00

Oh, it was.

Annette Bakker, Ph.D. 13:01

Yep, the same, the same inoperable plexiform neurofibroma for NF1. But what people didn't know is that this is in fact a pretty big population. So NF is about one in two and one NF1 is about two and one in two and a half thousand. And then 50% of those develop plexiform neurofibromas. And then there is a fraction of those that are inoperable, but not that it's not that small of a fraction because these tumors grow on nerves. So if you remove the tumor, you actually damage the nerve. So that's why a lot of companies are looking at treatments rather than at surgery.

Ben Comer 13:35

So if it wasn't licensed, how was it moved over into Springworks for development? Can you talk to me about that process?

Annette Bakker, Ph.D. 13:43

Yeah, so there were a couple of, I called in believers like Frida Lewis Hall, like Bain Capital or Bimet. There were a couple of investors. Then their life arc was in there as well. They raised about $130 million. They spun out from Pfizer the company called Springworks. Management team was appointed, so Sakip um was appointed, and then they developed the drug, two drugs, actually, not only Murda Metanip, but also a drug for desmoid tumors. Both were approved in spring of 2025, and then both drugs were acquired by uh by Merck, by the German Merck for $3.5 billion.

Ben Comer 14:22

Right. How was that investor syndicate brought together? You know, were you just kind of out there talking about this potential candidate? And, you know, how how did that how did that happen?

Annette Bakker, Ph.D. 14:33

No, at that time, no. So uh we were more, for example, we convened the clinicians to make sure that the clinical trial could be designed in a very short time. Um, we made sure that if there was anything that needed to be done to from towards the patients to get the patients into the trial and to really help publicize the trial, but we didn't really actively involve with the um investment community. What we're doing now is we're having regular calls with investors to kind of bring them into the financial pipeline, because on the one hand, and and Andrew can talk way better about this than I can, but I see a lot of money on one side of the equation, a lot of science on the other side of the equation, and a big gap in the middle. And now for the moment, the answer is very often venture philanthropy, but I'm like, no, we don't have that kind of money. So what what we're now really looking at is can we build a financial pipeline and say, we will bring it till here? Can you, the investment community, come a little bit towards the middle so that we can actually close the gap? So that is now uh we're building that um relationship with the investment community.

Ben Comer 15:47

And when you say bring it to here, where would that be? Is that preclinical? Is that you know a phase one trial? How far would you potentially progress it before you would need those investors to come in and and offer their support?

Annette Bakker, Ph.D. 15:59

So the usual answer is from the investment community is come back when you have phase two data, right? That is the that that's the standard answer. Yeah, and then I'm like, then I don't need you anymore because then everybody is gonna be interested in the in the in the asset, right? So I want I am trying, we're trying to push them a little bit out of their comfort zone and say, is there a step a step before that where you would be interested? Or could we invest in, for example, we have no problem investing in the preclinical work, um, especially when it comes to the shelved asset. We have the advantage that we have a lot of safety data and a lot of data on these assets already. So we can very quickly get into signal finding trials. So the question is, what is that minimum data set that you investors need to see for you to say, like, okay, we will start coming closer to the other part? But it's an it's still an uphill battle, I have to admit.

Ben Comer 16:58

Um, my I one other thing I wanted to ask about Springworks uh and the the Merck uh KGAA German Merck deal. Um, do you think that Springworks would have been able to successfully commercialize Gomekley if that deal hadn't materialized?

Annette Bakker, Ph.D. 17:15

Oh, they they did. The drug was approved by SpringWorks. Um it's a I I can't, I mean, I'm I'm gonna let it be.

Ben Comer 17:25

They started out to commercialize.

Annette Bakker, Ph.D. 17:26

Andrew is more in the financial world than I am, so I'm not I'm I'm I always say I'm a simple pharmacologist, right? So I'm not. But Andrew can definitely answer that question much better than I can.

Ben Comer 17:39

Yeah, Andrew, what do you think? I mean, I I think it was approved in February. Uh Merck announced its intentions, I believe, in March, the deal closed in July. Could could it have been a successful commercialization, you know, without a large, larger company, you know, with massive commercial networks coming in?

Andrew Lo, Ph.D. 17:59

Well, it would certainly have been more challenging to do that, but it has been done. Uh, for example, this company that I co-founded a few years ago, Bridge Bio Pharma, started out as a biotech company that's developing drugs in early stages. And once we got an approval, there was a there was some uh uh you know expectation that we would sell that drug to a big pharma, but Bridge Bio made the decision that they were gonna build a commercialization team. And now the drug is being licensed and marketed by Bridge Bio. So there are cases of biotech companies that have decided that they want to take on the challenge of commercialization. And I expect that Springworks could have done that if they decided that they wanted to pursue that

Why Shelved Assets Stay Shelved

Andrew Lo, Ph.D. 18:40

path.

Ben Comer 18:41

Right. Um, Andrew, I want to stick with you for a minute. Uh, repurposing abandoned development candidates is not a new idea. There have been some notable successes over the years. Viagra famously was initially developed as an anti-hypertensive uh thalidomide, even, you know, removed from the market for birth defects in the 1960s. Everyone knows that story. It was approved for multiple myeloma uh in 2006. Um, but that kind of repurposing doesn't happen as often as perhaps it could. I suspect it's for a number of the reasons that Annette already mentioned. But um, but why else? Why why is that not happening more actively when there are potentially thousands of these candidates languishing uh on the shelf?

Andrew Lo, Ph.D. 19:30

Well, I think there are several different reasons. Some of them are economic, some of them are psychological. Um, the psychological aspect is that investors are always interested in the shiny new thing. Right. And repurposed assets are not new. They're shelved. And so there's always this suspicion that, oh, maybe they're a shelf for a reason and you know they've got cobwebs and dust and all sorts of stuff, and we're not really interested in shelf things. Um, I think that the economic reasons are that uh because of the uh long time that some of them have been languishing, uh, there may be a concern that uh, you know, the the the the cost of restarting it is going to be uh you know quite a bit more significant than focusing on something that is already in the clinic and where you have more data and you don't have to worry about uh trying to round up all of the experts that have now dispersed when they were uh developing the drugs uh together. Um so it's a combination of those two uh reasons. But you know, as we've seen, there are great rewards to spending the time to sift through all of these candidates, and there are many, not just a few, but in my opinion, there are many diamonds in the rough. And what they need is some polishing and TLC, and uh, and then you get amazing uh results like O'Meckley.

Ben Comer 20:52

Yeah, I'm glad you mentioned the psychological piece because I have spoken to CEOs who even sort of recoiled from the word of repurposed because, you know, just for the reasons that that you mentioned, not because anything is wrong with the drug, because it's seen as perhaps something, there was some substantial reason uh aside from just you know portfolio strategy or or any other number of reasons that a company decides not to develop a drug. Um uh so that's interesting, but I want to ask you about the sifting through these shelved candidates. Um, is there clear visibility to these candidates that are sitting on the shelves uh across hundreds of companies? Is that information readily available?

Andrew Lo, Ph.D. 21:34

No, it's not. That's actually why we published our paper. In fact, it was at this session that Annette organized, and she very kindly invited me to participate, that ultimately, you know, we had to throw the gauntlet down and try to challenge the audience to do something. Um We had discussed the possibility of using the repurposing in a more effective way. And oh, a couple of questions emerged from the audience. About, you know, how is it possible to do that? Because, you know, biopharma companies that have shelved assets aren't always ready to disclose what those assets are. And so it just seemed like the challenges were insurmountable. And so that really frustrated me because, from my point of view, you just got to get started and do something instead of just saying that it's impossible. And so I proposed a very simple approach to identifying potential shelved assets. And the approach is to go to clinicaltrials.gov, that's a publicly available database. Anytime a clinical trial is launched in the United States with patients, they have to register at clinicaltrials.gov. So what if we looked at all the clinical trials that were launched and identify the ones that were not terminated for any reason, they're still open, but maybe has had no updates for a while. Those are the examples of potential shelved assets that you can start with. So that's a superset of all possible candidates. And then from there you can narrow it down by looking at press reports and looking at the FDA website for complete response letters and other sources of information. You can you cross-check and verify. And now with AI, we have the ability to do that at scale. So, you know, I said all this at the meeting. I said, look, you know, it's possible to do that, and here's the recipe for doing it. And I I was hoping that there'd be some entrepreneurial sorts that would uh stand up and say, Yeah, this is something that I'm willing to do. Nobody did that. And so that's when Annette said, Well, maybe we ought to do that. And, you know, I said, Well, MIT is certainly up for the challenge. We can uh use our analytical framework to be able to produce that list and then winnow it down. And and thanks to uh Annette and uh uh Frida and uh Tanisha, we were able to put together uh a great academic team. Um Richard Thacker and uh Josh Lerner and uh uh Mohab uh uh they were all wonderful in terms of being able to uh you know uh pitch in and put together, I think, a really compelling uh analysis.

Ben Comer 24:13

And I think is I think one of the benefits of mining clinicaltrials.gov is you're talking about products that at least have some amount of clinical data. You're you're not talking about preclinical drugs that have not yet reached the clinic. Because I I correct me if I'm wrong on this, but I I think something has to be included in clinicaltrial.gov only when it goes into human trials.

Finding Candidates And Incentives

Ben Comer 24:35

Is that correct?

Annette Bakker, Ph.D. 24:36

Yep.

Ben Comer 24:36

Yeah.

Annette Bakker, Ph.D. 24:37

Um maybe Ben, I I wanted to bring up one thing that um what I've observed in these shelved assets is in fact I've I've looked, there is a few types of shelf assets. The shelf asset could come from the fact that a company reorganizes their pipeline, changes strategy, and the asset gets on the shelf. Number two is there is a small biotech who has X molecules. Two molecules of those are of interest to the major company who acquires the biotech, the other one goes on the shelf. The third one is that there is a biotech that has a big strategic partnership with a major pharmaceutical company, and then years later the pharma company says, Oh, we don't want to develop this drug anymore, and the biotech gets the drug back, but can't do anything with it. And then the fourth one, which is the saddest of all, is that if a biotech has submitted IND, has submitted already some clinical data on the molecule, but the company goes belly up, then that molecule is kind of in limbo and nobody can access it. I think that those are kind of the four types of shelf assets that I have seen. I think one of the main challenges with pharma is that it's nobody's job in these big companies to exit shelf assets. Nobody's job. It is everybody business development is looking at in licensing the next GLP one, right? The CMO and the CSO are working on the active pipelines. Nobody it's nobody's job. So if we want to change that, it has to first become somebody's job in these companies. Because just to give you an idea of Springworks, to get Goldmeckli and the Desmoid Tumor Drug out of Pfizer, that required 200 volunteers. Because when the data is a little bit everywhere, the teams are not there anymore, are working on something else, that was that's the level. So that is not scalable. So that is why I think we really need to figure out incentives, whether it's regulatory incentives, whether it's financial incentives. Something has to, whether it's reputational incentives, but pharma has to benefit from making these shelved assets available. And I think that that is probably for the moment the biggest challenge. But our phase one was let us show what is the size of the opportunity. And now phase two is how are we going to do it, right?

Ben Comer 27:06

Right, exactly. And with in the case of Gomlecky, you you had this relationship with Frida Lewis Hall, which it seems like helped, you know, to move that drug, yeah, helped a lot to move that drug into Springworks. Let's say, let maybe a hypothetical, if you found another um asset that looked very promising, I identified through clinicaltrials.gov, maybe a company that you didn't have as close of a relationship with, what would be the approach like there? You know, what would you who who would you engage? What what would be your proposition?

Annette Bakker, Ph.D. 27:37

So that's exactly what we're doing. We have about 30, 40 assets for NF that are shelved and that we would like to get access to. And um it has been a little frustrating, I have to admit, that a lot, you know, you constantly talk and then you get kind of thrown over. It's like a hot potato, right? You are you, oh, I'm going to introduce you to this one and to this one. It's all very polite, but there is not a lot happening. And I wrote on OP-Ed a couple of months ago or last year, or maybe, about, you know, we need to find other Fridays because Frida was like a dynamo within Pfizer. She believed in it and she still does. And she could make it happen. And I'm sure that in all the big pharma companies or even the biotechs, there are other Fridays. We just need to find them. We need to find believers, I think, Ben. That's the most important.

Ben Comer 28:27

Believers and an advocate within the company. Yeah, that that makes a lot of sense.

Annette Bakker, Ph.D. 28:31

And I think an advocate who is high enough up that people would listen. You know?

Ben Comer 28:37

Right, right. And I've I've read recently uh about biopharma companies becoming a little more secretive about early development candidates just in terms of drug mechanism and target, uh, fearing perhaps that a Chinese biotech, you know, can outrace them to clinical trials. Is that a a fear that that you have to work against now? Are you are you sensing that or seeing you know a new layer of secrecy or no?

Annette Bakker, Ph.D. 29:05

Well, there's some companies that have been very secret about it, but then you avoid them. I think it's now a question, and in fact, Andrew had had a nice group of MIT MBA students that did the study for us, and it's a question of finding five assets, five, only five. And then we will have a win. And I think we just need a couple of wins. Because when I was talking at bio to some investors and to some people with money, they were all like, yeah, you know, Annette, one is nice. And I'm like, Yeah, I know two is better, but you know, so what we what we need is just a couple. And there is what is interesting is that we really see now a group of companies that are contacting us that are that actually want to build their business around shelved assets. So there is a an energy, and there are some investors that are that believe that are believing in it. And there are a few companies now, right? Like Ceravel, Pixie's oncology. Uh, there's just a company spun out of BMS, B Medicine. So there are there are more and more of these examples, but it's still one-offs. And I think that is where Andrew had the great idea that

Marketplace Matchmaking And AI Tools

Annette Bakker, Ph.D. 30:15

instead of recreating these one-offs, can we do something a little bit bigger?

Ben Comer 30:20

Yeah, Andrew, talk to me a little bit about you know what what you're creating to help enable this, you know, repurposing model to work. I I read with interest, you know, uh, about a potential matchmaking, I mentioned in the introduction, a matchmaking mechanism that would potentially connect uh not just uh developers but but also investors. The the key aspect as we've been talking about to moving these drugs forward uh in clinical trials to approval. How does that work? How is that set up?

Andrew Lo, Ph.D. 30:53

Well, first of all, the starting point for this is the research paper that Annette and I and our co-authors published recently. That was a kind of a proof of concept for how something might like this might work. And the proof of concept was a collection of data that we developed from industry sources. We have two very productive partnerships with Sightline and Cortellus, where they give us access to their data and we cross-check that with clinicaltries.gov, and we got a data set of something like several hundred potential candidates that were shelved. And they had a variety of different characteristics. So what we propose to do, and this is something where industry really needs to collaborate with us, um, is to put together a database of verified shelved assets. What we identified were potential candidates that were shelved, but we don't know for a fact that they've been shelved. They could be data errors, it could be uh discontinued for particular efficacy reasons or safety reasons. We don't know for sure. But the idea is that because we were getting several hundred hits, the universe of potential candidates was large enough that we thought, well, you know what, this could be a really valuable database if we vetted the particular candidates and then worked with their owners to provide a forum for various different parties to bring their own capital to bear on these candidates and to breathe new life into them. So the next step that we're working on right now is to talk to these industry leaders at the biotech and pharma companies to allow us to create this repository and then we'll connect that with various different investor groups as well as patient advocacy groups because we see a lot of energy and interest and funding to be able to take these shelved assets and bring them back into the clinic. So that's the idea. And we don't know whether it's going to work, but we've been getting a lot of interest from all different parties. And, you know, the clearly the economics work. You know, the GoldMekli is just one example, there are many others, of shelved assets that have become blockbuster drugs. So we believe that there are enough economic incentives if we can create the right informational structure to be able to bring buyer and seller together.

Ben Comer 33:18

Annette, uh, as a former oncology drug researcher and developer at both JJ and Sienna Biotech, what what putting yourself in those shoes again, what do you see as the biggest hurdle to for collaboration, for accessing and successfully developing some of these candidates?

Annette Bakker, Ph.D. 33:39

Well, I think first of all, the challenge is to find all the data and to make sure that all the data is in one place and the data is well organized and that in fact, because you know what, drug is not abandoned from one day to the other, right? There is a reduced interest, the data is collected a little bit less stringent, blah, blah. The people that are working on the projects are being either leave the company or are being reallocated to other projects. So there is there, you need to create an incentive. I always say we need to create something that people are able to do what needs to be done. And instead of blaming companies for not doing this, I think we have to give them a good reason to do it. And whether that is financial, whether that is regulatory, for example, one of the advantages that I saw for shelved assets in NF is because NF is a rare disease. So if we can get some of these assets approved for rare disease, there is a patent life extension, market exclusivity. So there isn't, there isn't motivation to invest. But I think companies are under a huge pressure now to make new medicines very quickly. And we just have to create an environment where they can they are able to do the right thing.

Ben Comer 34:59

Do you get the sense that information technologies are are improving at pharma companies? And I ask this because, and you mentioned it earlier, data in various silos uh across an organization have heard that for years as being a real hurdle uh to doing all sorts of things. Um, is that improving or is it still a pretty tall order to ask a company to track down, you know, every piece of data associated with one of these shelved assets?

Annette Bakker, Ph.D. 35:25

Well, the the I I don't want to speak for every single farm under the ship. Yeah, sure. What I have seen with Springworks was that it was a pretty big lift to get all the data that was necessary to bring this drug to DNF trials. So I guess Pfizer is not the only one. I guess the other companies probably have similar challenges. Um but yeah, is AI going to help here? I don't know, but today AI looks like being the solution for everything. So I think we just have to be, you know, it's it's a bit an a scary moment where everybody has the answer to every problem is AI. So I'm not sure if this is AI, but there is a lot of rigor in making sure that we get the right drugs that are efficacious but also safe to the patients. Um, because you know, it there was a little bit of it at a certain point in time, especially in the repurposing space, which are drugs that are approved already and that are reutilized for another disease where there is no real money to be made. But there was a little bit of a concept that the next repurposed or repositioned drug was going to be found between the ham and the cheese, and then you just give it to the patient. And I think we have to be very realistic that this is a costly operation to bring these drugs actually through the clinics to a commercial success.

Ben Comer 36:46

Yeah, absolutely. And and Andrew, you you referenced AI uh earlier in the conversation. Uh, what what is the AI component to this model? How do you how do you envision AI uh working to either help you know identify assets, matching them to developers or companies or investors, perhaps?

Andrew Lo, Ph.D. 37:06

AI has played a huge role. It is a total game changer. And let me explain. So, first of all, I do think that IT is improving across the entire biopharma industry. But to Annette's point, it fits and starts. So there are some companies that are ahead of the game. Other companies they may be a little lagging behind, but across the board, all of the various different biopharma companies are improving their IT infrastructure because they have to. They have to keep up. But on top of that, you add AI to the elements of you you add AI to this picture, and a lot of things are different. I'll give you a very simple example. For our analysis in the publication, normally we would have had to sift through literally tens of thousands of entries to see whether or not a particular shelved asset was shelved because of uh efficacy or safety or because of economic reasons. And we were able to do that literally ten times faster than we would have by hand, thanks to the various different large language models. Now they weren't perfect, and you had to make sure that there wasn't any kind of hallucination. That did take some time. But the fact is that these LLMs are improving literally every day. And so tasks that would have taken months or years to sift through large amounts of data can now be done in literally minutes. And that's really exciting. But you know, that's a very small illustration. Let me give you a much more ambitious one. So repurposed assets has multiple components, but one of the most important is medical underpinnings that would suggest that a particular repurposed asset could be relevant for a specific disease.

Ben Comer 38:51

Right.

Andrew Lo, Ph.D. 38:52

Now you can have LLM sift through the medical literature, patents, databases, the data records of these shelved assets up until the time that they were shelved, and basically piece together a hypothesis that says, you know, this compound might actually be effective for NF. Whereas before, we would have had to have a bunch of scientists pouring over large amounts of paperwork to be able to sift through that. Now it can be done by a few chatbots working in collaboration with scientific experts. So AI is not an answer by itself, but what it does is leverage the expertise that Annette and her group of scientists can do in days, what would have taken literally years. That's the exciting thing about repurposing. It's happening now at the perfect time where you've got scientific breakthroughs, the medical understanding that we didn't have before, and on top of that, the ability to sift through humongous amounts of literature that we can now correlate with all of the various different scientific hypotheses that we were hoping

FDA Pathways, PRVs, And What’s Next

Andrew Lo, Ph.D. 39:55

to prove.

Ben Comer 39:57

Great. Um and Ed, I wanted to ask you about um the FDA. And you know, you you alluded to some um potential regulatory incentives or or or policy changes that could help to move this whole area forward. I'm curious about what you make of the the FDA's proposed uh plausible mechanism framework. I mean, is that a program that could potentially help bring NF treatments to market faster, in your opinion?

Annette Bakker, Ph.D. 40:25

Um, yeah, there's a lot of um FDA initiatives now that are I mean, we are being invited to the August 5th meeting to talk about drug repurposing. I think that is more about the drug repurposing of approved drugs, but I think there is an enormous energy now in the field of yes, there is such a potential of drugs that are being shelved that could help patients and that could have commercial success. Why don't we look at them and why don't we see what can be done to actually bring these drugs to the patients? Um, one of the things that I was talking to the FDA about is, for example, the companies that the the the to me, the saddest and maybe the lowest hanging fruit are those companies that have submitted information to the FDA and then went belly up, and that information is at the FDA, it is still proprietary information from the company that doesn't exist anymore. So, what can we do together with the FDA to actually make that information available to maybe a new innovator or a new entrepreneur, right? So that's one one way.

Ben Comer 41:37

Is there any process to do that right now?

Annette Bakker, Ph.D. 41:40

No.

Ben Comer 41:41

Wow, okay.

Annette Bakker, Ph.D. 41:42

For the moment that data is there, and and you know the FDA has only limited liberty to make that data available to the same thing.

Ben Comer 41:48

Yeah, right.

Annette Bakker, Ph.D. 41:49

No, that data is there. So the so the and I think the maybe the lowest hanging fruit for the moment to get access to shelved acids is to work with biotics. We still need to inject the money, because that of course they don't have the money to develop the drug, but they're honestly easier to work with to make the assets available at at cost or at no cost, but they are sometimes easier to work with than the major pharmaceutical companies because of course the the level of you know bureaucracy of the major pharmacompanies is is more complicated than in a biotech.

Ben Comer 42:27

Yeah, right. Uh Andrew, I wanted to ask you about the this plausible mechanism, which I think is designed specifically for very extremely ultra-rare conditions, even you know, four or five patients or or even single patients in some cases. My my question is, is do you think, Andrew, that there's a a plausible business model for drugs that are only treating a handful of individuals? I mean, is there uh a justifiable to shareholders, uh ROI, you know, for for that kind of work?

Andrew Lo, Ph.D. 43:00

I think there is. And that was actually the topic of conversation at the session that Annette organized. What I proposed was a business model for ultra-rare diseases that's built around the priority review voucher. Now, as we know, the PRV has been reauthorized until 2029. Yes. And the hope is that it will actually be written into law permanently as opposed to sunsetting after a few years. I can tell you right now that with the PRVs, there is a very credible economic case for a completely private sector-based solution where investors will earn a very, very reasonable rate of return, uh even though you're only focusing on a very small number of patients. In fact, I made the argument at this conference that even if you never charged a single dime for the drug after approval, even if you gave it away for free, investors would still make a very handsome return just from the sale. Of the PRVs. So you can organize a portfolio of multiple ultra-rare disease therapeutics and not worry about selling any of them and still make a profit for investors and get these therapies approved thanks to the PRV program. And that's something I believe we're going to see over the course of the next few years. Now, without the PRV, it is more challenging, but even in those cases, if you create a blended portfolio of ultra-rare, rare, not so rare drugs and put them into a portfolio, all of those drugs can get funded and earn a very, very attractive rate of return for investors. So, you know, I think we just need to be creative about how we approach the financing. It's not business as usual. You do have to do a little bit more work, but that work is very rewarding in being able to generate cures for patients that would otherwise have no other alternatives. And I'm I'm very confident that there are many different business models we are not using, we are not exploiting, which could actually help patients.

Ben Comer 45:06

Absolutely. And just on the priority review voucher, I think listeners to this show probably know what that is, but I'll just note that sales of priority review vouchers are north of $100 million now. I've seen some uh touch almost uh over $150 million uh recently. So uh and that's no strings attached, you know, real real cash uh for companies that develop these ultra-rare drugs and and receive approval. So that that makes a lot of sense. Um I wanted to ask next if if you think this model can extend beyond NF and to other other cancer types or even beyond cancer and into other diseases, rare or or not so rare?

Andrew Lo, Ph.D. 45:50

I think it applies to everything under the sun. So uh we're seeing it operate very well in areas like NF and other ultra-rare conditions, but it can apply to any kind of disease. And certainly in the more common diseases, it applies in even greater force. You can have some very, very attractive returns uh if you're able to scale the portfolio to have this multiple shots on goal approach. Bridge Bio is an example, Royvant is another example, PureTech is a third example. There's probably 20 or 30 portfolio company-like structures in the industry right now. And I would argue that for a number of different conditions, this would be a very effective model in reducing risk, increasing the expected reward to investors, and bringing therapies to patients a lot faster.

Annette Bakker, Ph.D. 46:41

I just I think of course I completely agree with Andrew, but there is one additional thing is that sometimes rare is treated as rare, right? You don't talk about it, nobody knows. But rare is actually the entry point to precision medicine. And so I think if we profile it like that, I still remember when I was a kid, there was a kid in my class with leukemia, she passed away. Today we have a much higher really cure rate of leukemias because it's 114 different leukemias. So I think it's we're going towards a new system. And that is why I love working with Andrew because he's constantly thinking creatively about how to finance that. Because of course, the GLP $140 billion market, that can go with your cookie cutter business model, right? But once you get into this kind of rare disease, you get into precision medicine. Now you need creative, very smart people that are in the finance world. And that's that's what excites me working with with Andrew and the team.

Ben Comer 47:44

And Annette, I think you mentioned, uh, or maybe Andrew did, um, there are uh a bunch of other re um shelved assets that are are have currently moved into development through your efforts. Um can you tell me about maybe one or two that you're particularly excited about?

Annette Bakker, Ph.D. 48:01

So we did one, but that was really a repurposed drug. That was a drug, um that was a drug from Takeda, Brigatinip, which was approved for alk mutant lung cancer. And we demonstrated in our NF2 preclinic, NF2 related tronomatosis pre-clinical portfolio, we which was a huge collaboration with uh between about 20 academic labs. We demonstrated that that drug suddenly popped up as an active candidate. We nobody understood because there was no ALK in NF2. Um, so then there was a whole proteomic experiment that was done in Indiana, and it came out that in fact this drug was inhibiting an NF2 relevant target. And then it went into, in fact, Takeda then co-funded the first platform basket trial for NF2. Um and what is what is interesting now is that last this year, actually, the drug was included in the NCCN guidelines. So it but then here again, this drug has been very effective. It's very hard for a company to justify if you have the drug already approved for non-small lung cancer, which is a huge market. Companies are not going to put in the effort to also get it approved for a rare condition with 25,000 patients in the US, right?

Ben Comer 49:19

So additional risk, right?

Annette Bakker, Ph.D. 49:21

It's an it's not only an additional risk, but there is no real commercial it's hard to say, of course, from where I'm sitting and facing these patients, but commercially it's not so it's not the best way of using their money to um

Two Year Goals And Closing

Annette Bakker, Ph.D. 49:35

to get these drugs approved.

Ben Comer 49:38

Um a final question uh for both of you. Uh what what goals do you have? Uh um, let's say for the next two years, if we were talking in July 2028, um what what do you hope to see? What do you hope to have accomplished uh in in liberating these uh these stranded assets?

Andrew Lo, Ph.D. 49:59

Annette, do you want to start or should I?

Annette Bakker, Ph.D. 50:01

You can start.

Andrew Lo, Ph.D. 50:04

What I would like to see is an active marketplace where shelved assets are routinely listed. Kind of like on the NASDAQ or New York Stock Exchange. And individuals, organizations, patient advocacy groups that are looking to develop those assets would be able to download information from the website and be able to decide whether or not they wanted a bid on the asset. And ultimately, on a regular basis, shelved assets that are listed are going to be taken into the clinic and ultimately approved, kind of like eBay, eBay for uh drug candidates. That that's what I would like to see.

Ben Comer 50:45

And there there's uh, yeah, just sorry, sorry, before you before you um answer Annette, uh there's no uh correct me if I'm wrong on this, but there's no strong business case not to make this information available about uh these these assets that that are stranded, right? It's more the fact that there's not someone who's tasked to do that. Someone's having to do it over on the side of their desk. I mean, is that the main issue? Or or is there, you know, a business reason where a company might be hesitant to reveal an asset that they halted development on?

Andrew Lo, Ph.D. 51:18

You know, I think that in certain isolated cases, there may be strategic reasons where they want to hold off on developing an asset. But by and large, I suspect most of it is really just not enough time and much more pressing priorities. You know, biotech and pharma companies, uh, they they they, you know, some of the even the most well-funded of these companies uh are still deploying capital carefully and making sure that they make the best use of their capital uh in order to discharge their fiduciary duties to their shareholders. And so they want to be able to develop the drugs that they think are gonna be the most important and drugs that may not be their primary focus or may have some complexity that needs to be addressed, they simply put that on a back burner. And once they do, it's very hard to get back and try to understand exactly what the nuances are of each of the different cases. So, you know, it's kind of like your attic. You know, when was the last time you went up to your attic and cleaned out your stuff from junior high school? Um probably never. So uh it it's if the idea is to bring some light, you know, into these dark corners of uh, you know, these various different shelves. And um I think that by lowering the the threshold of energy needed to activate one of these shelved programs, that is what we believe uh can be done quite effectively with the um uh the process that we're hoping to implement.

Ben Comer 52:43

Excellent. And sorry, back to you, Annette, uh what you hope to accomplish, you know, two years from now.

Annette Bakker, Ph.D. 52:49

So the first thing I would like to accomplish is that the people understand that the listing of the shelved asset is not meant to be a blaming exercise, that it's meant to be a positive thing both commercially and medically, right? So that is the first thing. Then the second thing that I would like to accomplish is that we get the companies to the table to actually openly have these conversations and connect the shelved assets that we have identified, that they are connected to the diseases that could benefit from them. That we really build almost like a like a connector map that we say, like, okay, these 20 assets are good for Alzheimer's, these 50 assets are great for NF, these 20 are great for, and that we build like a connector map. And then the last thing I would love to have, of course I want to have 20 drugs for NF, right? That we can get these assets through the pipeline quickly and that we can actually really bring drugs to the market for our NF patients. But I think there is a bigger opportunity where it just becomes normal. It's like it becomes like recycling plastic, right? It's not anymore a thing that it's like there is this one or two or three exceptional people that are willing to jump in, that it just becomes a normal, a normal habit thing. Like, okay, we've invested this, let's make these assets available to for-profit, to non-profit. Uh it doesn't matter as long as they don't get lost.

Ben Comer 54:19

Do you think that can happen in two years?

Annette Bakker, Ph.D. 54:22

Um, I think with the team that we have now, we are all getting older and I think we're all getting equally impatient. So yeah, I think there is a there is an a resilience in an energy that making it the fully normal thing, maybe not yet, but creating sufficient energy that people are getting jealous. That that is, I think, absolutely accomplishable. I like I like jealous people who joined crazy ideas.

Ben Comer 54:54

Uh Annette and Andrew, thank you so much for coming on the show.

Annette Bakker, Ph.D. 54:58

Thank you for having us.

Ben Comer 55:00

We've been speaking with Andrew Lowe, PhD, professor of finance at MIT Sloan School of Management and Director of the Laboratory for Financial Engineering, and Annette Bacher, PhD, CEO at the Children's Tumor Foundation. I'm Ben Comer, and you've just listened to the Business of Biotech. Find us and subscribe anywhere you listen to podcasts, and be sure to check out our weekly video cast of these conversations every Monday under the Business of Biotech tab at life scienceleader.com. We'll see you next week. Thanks as always for listening.