Summary

Matt De Silva, Founder & Executive Chairman of Notable Labs, sits down with Neil to discuss his company’s predictive precision medicine platform and its move to leverage its platform to develop its own precision cancer therapies.

Available on SoundCloud, here.

Available on iTunes, here.

Transcript 

Danny Levine (Producer)
Yeah, we've got Matt De Silva on the show today for listeners not familiar with Matt. Who is he?

Neil Littman (Host)
So Matt is the founder of Notable labs. He was the company's first CEO during the first couple of years of the company's existence, as they were developing their core technology platform, which we'll get into in detail during our conversation. He led the company through its $40 million series B financing. As is often cases, companies grow and mature. Matt decide to step aside as CEO and he's now currently chairman still heavily involved in strategic level decisions for the company that and Matt and I have known each other for a half a dozen years. At this point, I, I first met Matt. When I got interested in this idea of the increasing convergence of technology with biology, I was doing a deep dive on the sector for bio verge, as were looking to make investments. I met Matt who had just started known Paul at the time.

Danny Levine (Producer)
Well, Notable is a Bioverge portfolio company. Why did invest in Notable?

Neil Littman (Host)
Yeah. I'll let Matt get into the founding story, but long story short notables, that was actually the very first investment that I ever made through by verge. We've been supporting the company ever since through multiple financing rounds. I even went in-house and helped run business development for a period of time at Notable leading up to the series B. I think for me, and for Bioworks, it was a, it was an early company in this space that has now become known as tech bio, but it was very clear, Matt's passion, you're here, the origin story, which I won't spoil, but, he was just, he was so, so passionate, so dedicated, really trying to solve a problem. He was coming at it from a, a or Fabiano approach and that he was an industry outsider, right? He didn't come from the healthcare world. He wasn't a PhD trained scientist.

Neil Littman (Host)
He was just trying to solve a fundamental problem. I thought that the platform that they created in this, it was very early at the time we made our first investment, but it was this bringing together at that time stem cell biologists and, cancer biologists with software and hardware engineers to build this automated high throughput platform. It was just the exact arc type of company that I was looking for at the time. And, I think the stars just aligned, they were just getting off the ground by verge was just getting off the ground. I could see the potential right there. Wasn't a ton of data for us to dig into. At that point, there was enough to say that the technology looked promising, but they didn't have any of the validation studies that Matt and I will talk about. But I saw the promise, right? If you could, if you could a priori predict how a patient would respond to a drug before ever giving that patient a drug.

Neil Littman (Host)
I mean, that is quite literally the holy grail of drug development. I could see the potential and I could see where they were heading. So, it was an early stage bet, but it's been really amazing to see how the company has grown a mature since our first investment. It's really incredible to see where they are today with our own pipeline of drugs.

Danny Levine (Producer)
You and I have been steeped in this world of precision medicine long enough that it was known as personalized my son when we first came across it, it's difficult to remember that when people are diagnosed with cancer today, they're not necessarily treated in this new world of medicine that they're still in a very traditional approach to cancer therapies. Where do you think we are in seeing precision medicine take called and making it the standard?

Neil Littman (Host)
Yeah, I mean, I, I still think we're relatively in the early days, I, I believe it was Gleevec, that was the first, oncology drug that was approved as a targeted therapy. That was in 2001, there's, since then, there's been a lot of new approvals for drugs based on different genetic mutations, but even based on a genetic mutation, that drug does not necessarily work for a hundred percent of those patients. In fact, it works for far less than that. Even when there's a genetic mutation that the drug targeted to that mutation it's still is not necessarily effective for those patients. So that is certainly precision based, right? A drug is targeting a specific genetic mutation for example, but it might not be personalized in the sense that drug is not necessarily going to work for that individual patient. And so I think what Notable is doing is, is really another way to go about, you know, targeting really more personalized medicine where you're actually trying to predict if a given patient will respond to a particular drug.

Neil Littman (Host)
So to me, it's sort of the next evolution, you know, the, the technology that Notable uses is called, you know, phenotypic or functional screening. That's kind of been around for a long time, but they have developed a whole suite of technology around that. I think, whereas there are a lot of ideas that didn't work in the past. I think the time is now great for this concept of what Notable is doing. I think they've done a lot to validate the technology and the platform. I'm really excited to answer your question. I think it's still really early days. And I think a technology like what Notable is developing, I think would be hugely powerful to sort of advance the whole field.

Danny Levine (Producer)
Well, if you're all set,

Neil Littman (Host)
Let's do it, Danny. Thanks for joining us today. I'm thrilled to have you on the show with us.

Matt De Silva (Guest)
Really nice to be here. Thanks Neil.

Neil Littman (Host)
So today we're going to talk about precision medicine, Notable labs, predictive platform technology, and it's moved to develop a pipeline of novel therapeutics. However, before we dive into the details, I love to start with the origin story of Notable. It's, it's a story that I can only describe as sad and yet beautiful at the same time. So, you know, Matt for our listeners, can you provide a little context about the why as in why you started Notable.

Matt De Silva (Guest)
Absolutely. It's always, I think, useful to start with that as background, because it informs everything that we're doing today at notable. I'm not the typical life sciences founder. My background is actually in behavioral economics and I was trading currencies as a portfolio manager at Peter Teal's hedge fund. In California, when my dad was diagnosed with three brain tumors called glioblastoma multi-format, which is a very aggressive disease. Unfortunately, because he had these three separate tumors, when I made a list of every clinical trial around the world, that we might be able to enroll him in. It turned out that many of those were, were clinical trials that were not available to him because of those multiple tumors. So they were called an exclusion criteria. What was left was all of these different drugs that were being repurposed from other diseases. Some of those were, treatments for alcoholism.

Matt De Silva (Guest)
Very kind off the beaten path as a cancer therapeutic and others were maybe from another type of cancer, but for brain tumors, like a breast cancer drug. Those were available as potential treatments that a physician could prescribe for my dad. We could use a combination of those different drugs, but there was no method to know which of them would make sense for him. Right. That's really where the precision part of medicine is going to come in to this narrative. And, and we had his tumor sequenced, right? We tried to look for mutations that might give us clues as to which drugs would be active for him. We had a company that put part of his cancer cells into mice that had compromised immune systems. You can run a clinical trial in the mice. We tried to have an academic grow his cells in the lab so that we could test drugs on them.

Matt De Silva (Guest)
Unfortunately all of those different methods did not work, give us any additional treatment options. Ended up starting notable to take a lot of the advances that had been coming in academic labs and settings and, and apply those in my dad's case. So, began in the, in, in the summer of 2014, and we will, while were able to take my dad's sample and test different combinations of drugs on that sample, after growing it in the lab, by the time we had treatment found those treatment options for him, he had a fourth tumor that formed deep in his brain stem. We weren't able to, to turn that into treatment, but we decided to keep going with the company. That's, it's been very exciting to see the progress that we've made and, have his, his journey as part of that legacy.

Neil Littman (Host)
Yeah. I mean, I, I find it to be an incredibly powerful story and obviously starting to hear about the loss of your father, but I think in many ways, it's, it's inspirational and it's been amazing to follow the notable journey over the years and see how the technology started and how you thought about developing, a platform really with the patient in mind and where the platform is today. With that being said, let's start with what we even mean by precision medicine, how it's applied cancer. I think most of our listeners have heard for a long time about the promise of precision medicine, its ability to improve outcomes in a variety of different diseases, including cancer. Just so we're all on the same page. Let's, let's start with definitions. What does the term precision medicine mean to you?

Matt De Silva (Guest)
I think that we call it precision medicine is because what are we defining it as being different than, right? If it's precision medicine, what's non precision medicine, what's what is, and to me, that standard of care, right, that's basically saying, okay, you have been diagnosed with acute myeloid leukemia. Here are the standard chemotherapies that we're going to treat you with and treating all and giving all of the patients the same drug precision medicine approaches. Okay. We're going to take some kind of a technology that will tell us something about your disease and what drug might work for that disease. We're tailoring therapy for that specific patient. I think what gets even more exciting to me is actually the step beyond precision medicine, which is personalized medicine. Right. Okay. For your particular, cancer type and cells here is a drug that we think will work for you.

Matt De Silva (Guest)
Here is here's the probability that we think that is, that actually is going to work. Right. That's the predictive precision medicine part of notables story where it is today.

Neil Littman (Host)
Yeah. I, I definitely want to dive into that in a minute, but I do want to provide just a little more color around what you were saying, this idea of precision medicine versus in precision or imprecise medicine, which I think is largely the, as you said, the standard of care. Just a few stats for our listeners, right? The top 10 highest grossing drugs today only help between four and 25% of the people who take them. So imagine that. In other words, every day, there are millions of people who take medications that won't help them. The problem is we can't identify who will be a responder to that drug ahead of time or a priority. If you look at ICology specifically, right? This new wave of checkpoint inhibitors work for less than 12% of the patients that are prescribed for, right. Again, you can't predict who will respond or who won't respond to those checkpoint inhibitors.

Neil Littman (Host)
All of these patients are suffering all of these toxicities from these drugs that are not going to be effective for them. Of course you don't know that ahead of time. Ma I want to get into of the history of precision medicine. Bear with me here, just cause I want to go through the Gleevec story. Cause I, I believe that was the first instance of a targeted therapy being approved in oncology. It was approved for CML, I believe chronic myelogenous leukemia in 2001. And again, just for our listeners not to get too far in the weeds, but I think this helps provide some context about, you know, where we are today and what Notables is really working towards. So, Gleevec blocks the activity of the BCR ABL fusion protein that's based on a genetic alteration in what's known as the Philadelphia chromosome. That's just based on where the discovery took place.

Neil Littman (Host)
Gleevec is specifically targeted to and blocks the activity of the BCR, ABL fusion protein, since normal and healthy cells, don't express this protein they're not affected. Hence we have a highly targeted or precise drug that targets cancer cells, but not healthy cells. In terms of, that having the context of where we are today, where do you think we are in terms of the reality of precision or personalized medicine? I mean, how many more drugs are out there that are likely vac, is that the future of oncology and where do you take we are today in terms of that spectrum?

Matt De Silva (Guest)
Yeah, I mean, so just to bring it back to my experience with it as a field, when my dad's tumor was sequenced, were looking for different mutations so that we could target them with drugs made in the same image as Gleevec right. It absolutely, there's an era here that there's a before and an after, right. And, and the after Gleevec part, that's very exciting is that there's been other mutations in cancer. That happened in, in enough patients where the economics of making a drug that specifically targets the mutations vulnerability, has led to dozens of approvals. In my dad's case, his brain tumor had a mutation at ETFR and which is epidermal growth factor receptor. There are multiple drugs approved in lung cancer for EGFRs mutations. There's been a lot of effort in brain tumors to say, well, for patients with a brain tumor that has this mutation, let's use, a lung cancer drug that's already approved for that.

Matt De Silva (Guest)
That was actually one of the treatment options that my dad had. It's actually a direct kind of descendant of the Gleevec story. The problem is that even if you do have a mutation and you have a drug that targets that mutation, you don't know what the chances of the patient responding to that drug, right? Not all patients with an EGFR mutation will respond to any GFR inhibitor. Of course there are lots of patients out there that we still don't want to. We sequence their tumors, we don't find drugs that will target them. There's the two big kind of gaps that are, we're trying to, trying to address with the next generation of drugs and in oncology.

Neil Littman (Host)
And I think that's a great segue to now dive into Notables platform cause is taking a, a different approach, right? It's not a genetics or genomics based approach they have developed and you have developed a functional screening platform. Why don't we start with the core technology that underlies the platform. Describe how it's different from how like a Gleevec would work for example, or how a, a genomics based approach works.

Matt De Silva (Guest)
Absolutely. Yeah. So. Think of it as a brute force experiment, right? If we have a patients cancer cells outside of their body and we in a laboratory test drugs on those cells, they will, depending on their, all of the different mutations present in those cancer cells respond to various drugs, but it's a phenotypic system in that we are seeing how they respond, but we don't necessarily understand why we, we're not looking at the wiring here. We're looking at, okay, how did that cancer cell change after the drugs were applied? If that, if the patient then gets the drug after we have, or immediately after we have tested the drug in the lab, then we can see how accurate were. Did they respond to the drug or did they not respond to the drug? Over time it's a system that can get more accurate because when patients are not responding, but you thought they were going to respond based upon the lab test, you can update and add the additional measurements that were necessary to be accurate in predicting the next patient's response.

Neil Littman (Host)
Okay. Can you, I think an example would be helpful. Could you walk me through how the platform works? You start with a patient's blood sample, w w what happened? How, how do you go about figuring out what drug or what combinations is effective against that individual patient's blood sample? How do you then inform a physician? How do they then potentially make a clinical decision based on that walk me through that kind of flow, if you will.

Matt De Silva (Guest)
Sure. I'll give you a very specific example, which is we did a clinical trial with Dr. Peter Greenberg at Stanford university, who is one of the preeminent positions in a disease in blood cancer called myelodysplastic syndrome. We've also done a lot of work in leukemia with our, with our approach where, Dr. Greenberg and his team at Stanford would collect samples, either from the peripheral blood or from a bone marrow biopsy. Those would be sent to notables lab in foster city. We removed the red blood cells from the sample, but we keep the white blood cells because we want to measure the drugs activity on the cancer cells versus the activity of the healthy cells, because we want to look for drugs. This is kind of getting back to the precision definition of Gleevec. We want to look for drugs that are targeting the cancer cells, but sparing the healthy cells serve as a control for the experiment.

Matt De Silva (Guest)
We can take that same sample and test, thousands of different drugs and combinations of drugs on those cells by, by basically replicating the experiments over and over again, that's all done with a fully automated testing system that we built. We, we stitch all of that information together, create a report, and we rank order the drugs based upon how well this patient responded in the lab relative to all the other patients that we've tested previously. The drugs that were testing were selected by the Stanford physicians, right? They were saying, what are reasonable things that we might consider treating this patient with? It was in patients that had progressed on standard of care, so they didn't have any treatment options for them. If were able to find something that the doctors could prescribe, then that was going to be another treatment option for those patients. In the trial of which we've published, we published the study in 2020, and the journal called blood.

Matt De Silva (Guest)
We showed a 92% positive, predictive value. So in the lab, you predict response. The patient responds and an 82% negative predictive value of, if we don't think the patient's gonna respond, that then they don't respond. Right. All of that together gives us an accuracy overall of 85%, which were extremely excited to see for those patients.

Neil Littman (Host)
I want to talk about what that means, the negative and positive, predictive values. Before we do, I guess, two questions for you. One, you had mentioned you're now obviously focused on blood cancers. Notable started, as you had mentioned in glioblastoma, why the change from glioblastoma solid tumors to blood cancers?

Matt De Silva (Guest)
Yeah. In, in glioblastoma specifically, you get a very small amount of cells from the surgical suite because of how the tumor is physically removed. To do a large number of experiments, you have to grow those cells in the lab. That takes significant time that in the context of glioblastoma patients don't have. We ended up moving into blood cancers because when we get that tube of blood in the morning, we're testing the drugs on it in the afternoon, right? There's no period of time in between getting the sample on needing to grow it. We can do things ex-vivo where we test the, on the sample, right away. That reduces the chances that as you grow cells in the lab, that those cells change. We've been very focused on blood cancers as the proving ground, for this technology, because we get a large number of cells that we can do these experiments on directly from the peripheral blood or bone marrow.

Neil Littman (Host)
As often happens. One question leads to another. I have another question for you, based on that answer, you say testing drugs, what is the range of drugs that you're able to test through the platform and what I mean by that? What are the different types of modalities that you could focus on?

Matt De Silva (Guest)
Absolutely. Of course there are the bio verge portfolio and companies span across lots of different modalities, right? So it's a very relevant question. The majority of oncology drugs are still small molecules, right? Whether they're oral or Ivy, when we are testing in the lab, they're tested the same way. A lot of our work is definitely with small molecules, but we also can and do test antibody based drugs, which are used often in blood cancers. Those can be monoclonal antibodies, but they also can be, antibody, drug conjugates, or biospecifics w w then, then there's of course the car T cells have been very important, mostly in B blood cancers, but of course, there's quite a lot of interest and excitement around using those in myeloid leukemias as well. We are able to work with those in the lab setting. It's easier to do so with an allogeneic product and off the shelf car T cell, then the sample directly, then the car T cells directly made from the patient's on T-cells.

Matt De Silva (Guest)
We've been able to really work, frankly, across almost all therapeutic modalities. What's exciting to me is we can combine them so we can combine small molecules with antibody based, therapeutics, for example, in the same test tube and see how it does that combination compared to each of those drugs being tested on their own, in the same sample.

Neil Littman (Host)
Yeah. And, and I want to move to Notables pipeline momentarily, but I want to go back to this idea of a phenotypic screen or your functional based testing. You don't know, notable is not the first company to pursue this path it's been around in academia even longer than companies have been developing this type of technology in an attempt to predict patient response. Could you talk a little bit about, you know, how or why Notable is different than what has been done in the past?

Matt De Silva (Guest)
It's a very important question. And the history here is long. I mean, it really started off as early as the 1960s, but Pete in popularity and hype in the 1980s. We are definitely talking about a long history here. The issues back then were several fold. The first was the drugs themselves were relatively non-specific. This is pre Gleevec era, right? These are different chemotherapies that work in one way or another by poisoning cancer cells. They have a very narrow if the term is therapeutic window, but again, that's just a, a version of saying killing cancer cells relative to healthy cells. So the drugs were not as specific. Picking among all of the different chemotherapies is much harder to do in a lab based setting like this, then picking which patients are going to respond to a more specific therapy, like Gleevec, or like an EGF AR inhibitor, just to kind of keep the thread going from the prior part of the conversation.

Matt De Silva (Guest)
So the drugs have gotten more precise. That's a, that's a very important, why now for this approach. The conditions that you test the cells and are very important. If you have, if you don't have a biologically relevant setting to apply your drug, then you can create false positives and false negatives from the condition that those cells are in. Right. Our understanding of different ways of keeping cells alive outside of the body has had major advances in the last decade, mostly actually from the field of stem cell biology. Kind of applying stem cell biology to do a cancer context has been really important for, controlling the environment and making sure that you're not causing those cancer cells to change outside of the body consider, considerably, and then finally what you measure. Right? The tests in the past, we're very focused on measuring proliferation as a readout to predict response at the time, that was the thought of the driver of cancer, right.

Matt De Silva (Guest)
That, cancer cells really w it was a problem of out of control proliferation, but it turns out that there are many other things that drugs do and things that cancer cells do that are not just proliferation. It's really important to also be able to measure other things that would predict response in cancer, patients like apoptosis, or, viability, if it's an immunotherapy is an engaging the T-cell right. So, so being able to measure accurately how the drug works, biologically relevant setting with not just chemotherapies, but precision, targeted drugs all makes for a very different environment to apply this same idea today.

Neil Littman (Host)
Yeah. I think that the history is incredibly important and just to see where, how far things have progressed over the years. So I want to transition from the platform that Notable has really been focused on developing, you know, really since inception, you know, you've validated the platform with, you know, third party studies, you know, in partnership with Stanford and many others in November, Notable announced that they acquired the worldwide rights to Verlasso Tim from onco heroes, buyer sciences. This was a, a clinical stage product. I believe it was in phase three for acute myeloid leukemia. Before we get into the specifics of the drug. Can you, can you frame for our listeners, how you think about using the platform to develop a internal drug candidate?

Matt De Silva (Guest)
Absolutely. The last or tip is a precision medicine. It is, it is a drug that was designed to target as specific kindness. It's very much in the same vein as Gleevec right. This is a targeted therapy, not a chemotherapy. And it's target is called Palk one. So polo like kindness. One is the target. It hits that target, but it turns out that a patient's level or of PL K one expression has in prior clinical trials with this drug has not predicted whether or not they will respond to the Pok one inhibitor to Veloster tip. It's not a problem with the chemistry, again, that the drug was designed to hit a target that is relevant for cancer and has been shown to, to work in some patients. It's what it's missing is a way to, before you treat a patient predict who will respond or not to this drug, and that's where, notables platform and functional testing that we've been validating in the clinic for five years will be put to the test with this, with drug and others, that we will continuously be adding to our pipeline.

Matt De Silva (Guest)
If that makes sense. Like it's a precision medicine without a biomarker.

Neil Littman (Host)
Yeah. So I wouldn't be very clear here. This is, so this is really a way to select a patient based on the asset, right. Identify a priority before they're given the drug, whether a patient will respond or not respond. If they are identified to not be a non-responder, then they're not enrolled in the clinical trial. If they're identified to be a respondent or a responder through the essay, then you can enroll them in the clinical trial. Again, th the whole idea here is to increase the probability of running a successful human clinical trial, which is quite literally the holy grail of drug development, right? It's, it's what every VC is, what every pharma company is racing to do. If I'm understanding correctly, right, that is exactly what your platform, what you're asking is designed to do. Just to translate that one step further, right? This matters because if you can increase the probability of success, even by just a few percentage points, it has a massive impact on the value of the drug, right?

Neil Littman (Host)
From a risk adjusted perspective. Not only is it usually beneficial to patients, right? You're not giving patients a drug that won't work for them, but it's hugely valuable to us as investors, right? The companies like Notable, who are developing the drugs. I want to make sure that point is not lost on our listeners. Let's talk then about the development plan for Veloster tip. You've been licensed the product, where are you in terms of moving it at Ford and the, and the clinical development plan?

Matt De Silva (Guest)
Yeah, so the clinical development plan really is designed to take advantage of notables testing capabilities to target the right patients. You do that in, at a high level, in three phases. The, the first part of your, of your, your past is give all patients the drug in a small study and see how many of them respond and how many of them do not, but test their sample in notables lab, before they're enrolled in the trial to generate a cutoff point where you would predict response and non-response in future trials. The second stage is to, is the lock that as a diagnostic test, and then use the test as a pre enrollment criteria, and then only give the drug to those patients, which is what you articulated earlier. The final part is a pivotal study. That will be the final clinical trial for both approving the drug.

Matt De Silva (Guest)
And what's called a companion diagnostic. That that will be our test. After that approval, then every patient that is eligible with the given kind of cancer that we develop it in, and we'll focus here first and acute myeloid leukemia, we'll have a test run and notable, and then only those that are shown to be sensitive in that test will get the drug once it's on the market.

Neil Littman (Host)
I think it would be helpful just to talk about the current landscape for AML. There've been quite a few recent approvals over the last two or three years. Could you, could you talk about how effective some of the existing therapies are, what is currently, and if there is still an unmet medical need for a drug, like Verlasso tib, what is the prognosis for people today? I think that would just provide a little context for our listeners in terms of, w w w the, the competitive landscape.

Matt De Silva (Guest)
Yeah, absolutely. You're right in that there's been, I believe, eight recent approvals and acute myeloid leukemia. There's, there are many different therapeutic options, and it's very exciting because before that, there was decades of, of no new approvals in AML. The new options are very exciting, but they are not cures. They are, they are treatments that work only for a subset of patients. Many of the, the response rates are 20 or 30 or 40%. The duration of response, how long, those that the patients who do respond, benefit is often measured in months. Of course, to every therapy, there are exceptions and outliers, but even though we have many of these different drugs being approved, there are still many patients who relapse, or who don't respond initially to those therapies. So exciting times and AML, absolutely. Many more drugs are also being developed. The key we think is actually going to be in combining these drugs together, which again, is as what notables technology and strong suit has been in the clinic for the last five years.

Neil Littman (Host)
Matt, I want to talk about another drug that you recently announced a co-development deal for in partnership with cyclo med. Could you talk about that relationship and that drug?

Matt De Silva (Guest)
Absolutely. Cycle med is developing fossil Ciclopirox and in bladder cancer, and they've shown some, some exciting early clinical results with that drug, but there was also an interest in developing it in blood cancers. Notable entered into a co-development partnership where, we focus on developing the companion diagnostic that the test part for selecting patients for the drug and cycle med continues to focus on developing, the drug that they've created from scratch from the beginning. It's a great partnership because it lets notable see if we can predict responses for this particular drug in, in blood cancers and, publish those results and present those results. And, hopefully in the long-term create a companion diagnostic for selecting patients after approval, but they continue as a company to be able to develop the drug in bladder cancer, other solid tumor indications. So, we're excited to continue to add drugs to our, and diagnostics, our pipeline through partnerships that are co-development oriented.

Neil Littman (Host)
And I guess that's a good segue into the, the business model, you know, what the future holds for Notable, you know, w where do you see the company in five years? I mean, are you planning to become a a fully integrated biotech company? Are you planning to in-license additional, therapeutics, are you planning to stay focused within blood cancers? There a potential to expand to solid tumors? For example, I'll talk to me about where you think the company could be in the next three to five years.

Matt De Silva (Guest)
We are very excited about Veloster tip, because not only did it, has it been in a large number of leukemia patients? I mean, this has been in over a thousand patients in previous clinical trials. We know a lot about which types of cancers it's active in and which types of cancers it's not as active in. Some of those active cancers are solid tumors. We know have a very clear mandate as to, where to go build more tests in solid tumors. And, and we, as we've previously discussed, started as a solid tumor company and have been doing, testing for years, kind of waiting for the expansion back into solid tumors after proving that this works in blood cancers. We have a clear path forward there with Laster tab, but we will continue to build out our pipeline through additional in licenses and co-development partnerships. Eventually really using this actually for discovery and starting know programs from scratch ourselves.

Matt De Silva (Guest)
Long-term the business model for me is very exciting because once we have our first drugs approved with the companion diagnostic, if you're sending in that sample for patient to say, oh, are they going to respond or not to Veloster tip, we can use that same sample to test other drugs, right? Those might be drugs that are already on the market. They might be other drugs that are being developed by us or others, but the utility of that test goes up over time. We can start to figure out, right when the patients are diagnosed, which treatments they should receive, whether that's one drug or combination of drugs, we can do more testing after if those drugs stop working and find the next line of therapy and on and on again. That's really, that to me actually is very much the personalized medicine, holy grail, long-term vision of when you're diagnosed, here's the five drugs or combinations that are going to be most active for you.

Matt De Silva (Guest)
Here's the probability of you responding to each one and that's all backed up by, years and years of data. And it just gets better over time.

Neil Littman (Host)
And, and to me, that's also the fundamental platform aspect of the technology and feed over time. As you mentioned, you're ingesting more data, right? You're building a larger and larger data repository and moat, right? The moat is based on not just IP around specific drug, but there's a data moat as well. That as you said, just reinforces itself as you ingest more patient samples and the data set grows. I think that's hugely powerful, Matt, before I let you go, I wanna want to get your thoughts on this term that is relatively new in the industry, which is tech bio. And, and to, to me, this is the intersection of biology with technology, or, health and tech or whatever you want to call it. I started doing a deep dive in this area really when we first met him a bunch of years ago. I think the field has really exploded in recent years.

Neil Littman (Host)
There's a lot more, what you would call quote unquote tech, bio companies that have been started in the past, certainly three to four years, there's a lot more investor interest. W what does the term tech bio mean to you?

Matt De Silva (Guest)
It's is as a term that I think defines a, an industry and as a space that has been progressing for years. Right? I think were part of tech bio before it had a name, or went through Y Combinator in the second batch that they started, including life sciences companies. The first batch had Ginko. The second batch had us and Adam Wise and chassis, and, kind of has grown considerably since then. There's been dedicated venture capital funds that I think really have focused and capitalize on this space. I mean, I'll give you a more personal definition because, notable was started like a lot of tech companies where you have somebody who has a problem, right. In this case, it was, how do I find a treatment option for my dad? What's the technology that I will use to solve that problem. Right. The fact that's in biology, it has really defined us as a tech bio company, right.

Matt De Silva (Guest)
We built the technology and we applied it to this particular field. We've always had a tech startup mindset that blends, biology and engineering, because we started with that focus on patient need. And, and how do we solve it? How do we achieve product market fit? Many of those concepts that are coming from being in the bay area and being in the, the tech startup ecosystem, how do you apply that to a life sciences company, I think was the seeds that led to enough companies taking that approach, that it needed a name. And, and, I think kudos to, I think Vos Bailey's coining, and it's, it's been great to see now more and more kind of thesis about the various things that led to tech, bio driving it forward. I think another thing that is really important about it is that there are, many of these companies are founder led and very motivated to solve problems of biology with a playbook from tech startups.

Neil Littman (Host)
Yeah. I think it's, it's really fascinating. And, not only the playbook, but you had mentioned this, I mean, I, I love the cross-disciplinary team approach of these types of companies, right. It's, it's cancer biologists, it's stem cell biologists, it's chemists and it's software and hardware engineers. Right. You're merging these different disciplines, and I think there are, creating, immensely valuable businesses in doing so, because it brings together just different ways of thinking about the world.

Matt De Silva (Guest)
Yeah. I mean, it's great to see scientists learn to code, and it's great to see, engineers learn how to design experiments, but that is naturally what happens in the culture of these companies.

Neil Littman (Host)
Yeah. And, and, the, the culture is so important, so Matt, I think we could probably talk for the next four days straight about this stuff, but I want to be cognizant of your, and wrap up and just say a huge thank you for your time and for joining me on the show today.

Danny Levine (Producer)
Absolutely. Thanks so much now. Well, now, what did you think? I.

Neil Littman (Host)
Think that was a really great discussion. We were fairly narrowly focused on precision medicine within the oncology field, but I think it was a pretty wide ranging discussion within that. We got into the weeds on a lot of these topics, but I think, as you heard Matt say, I mean, I think the time is now right for this type of functional based approach. It's been around, as Matt said, like in the early days, since like the 1960s, there was a lot of interest in the 1980s. There were a lot of technical reasons why it didn't take hold. It sounds like a lot of those issues have now potentially been solved. I think the power of the predictive nature of the platform could be game changing. What I really love about what notable has done over the years is they really spent the first part of their life validating the platform.

Neil Littman (Host)
Not going out and trying to develop a drug, they really had third party validation that the platform is predictive. Both. You heard Matt mentioned the positive and negative predictive values of the platform. So, if patients respond or if patients don't respond and they have third party published data from Stanford and from others that validates the platform now, how is that going to translate into clinical trials and translate into for last year tab and their own pipeline? I think, remains to be seen, but, I'm incredibly excited about what they've done and, and the direction they're going and being able to apply that predictive platform to a specific drug that they now own.

Danny Levine (Producer)
It surprising that they've moved into therapeutics here? Does it say something about the economics of diagnostics versus therapeutics?

Neil Littman (Host)
Oh, yeah. That's, I mean, that's a really good question. I mean, there's, there's I think no question that the, the, the economics around developing a proprietary novel drug far exceed those around developing a diagnostic. So I'm not at all surprised the, the direction Notable is heading here in terms of, okay, let's build a platform, let's validate the platform. Let's, in-license a drug that we own use the platform to develop that drug, get it approved, and then let's, rinse and repeat, and let's do that for future drugs. Right. So let's really build a platform here. It's not, as, it's not a single asset plan, it's a initial lead asset to prove the viability that the platform works and to use as a case study to say, Hey, this worked for velocity. This can work for, X, Y drugs. I, I think it's not surprising at all to me that they're moving into this direction cause there's a lot more value to be captured in developing drugs.

Neil Littman (Host)
It makes a lot of sense to come to combine the companion diagnostic with, with ownership of a, a therapeutic,

Danny Levine (Producer)
How might having a predictive platform change drug development in terms of really improving rates for success and from an economic point of view, how significant could that be?

Neil Littman (Host)
Oh, it's usually significant, right? I mean, if you think about all the time and cost savings and, and patient lives, you could potentially save, if you could identify which patients are more likely to respond to a given therapy. So, I mean, first of all, you could have smaller numbers of trials in terms of patients, you could enroll less patients in a trial, if you can identify who is likely to respond versus not respond, right. You could cut trial sizes in half let's say in many instances, so you could run trials much faster. That would obviously save a tremendous amount of costs in doing so you could get drugs to market a lot faster. And, I, I, I, you heard me talking about the idea, the net present value and how that accrues a tremendous value to the drug developer and to the investor to invest in that company and that therapy.

Neil Littman (Host)
But, at the end of the day, it really beneficial to patients, right? If you can tell a patient that they're not going to respond to a particular drug and save them all of the associated toxicities of being on, enrolled in that trial and being on that drug, right. They could, they could go on and find another drug that may work better for them. So, it's just, you almost can't measure how valuable this type of thing would be in terms of, saving patient lives and really helping with the quality of life for patients. There's the economic argument, which is, I just walked through, which I think is hugely powerful.

Danny Levine (Producer)
Drug companies, always looking to give themselves some edge to improve success rates or avoid clinical trial failures. Are you seeing this as kind of a routine approach to drug development, to use some pre enrollment screen like this? Do you think it will become a, a standard approach for drug developers in the future?

Neil Littman (Host)
I mean, I'd love to see it become a standard approach in the future, right. I mean, as as I think a lot of us know, right. Th the biggest value inflection point when developing drugs is in, human proof of concept data, right? That's typically, post phase two, right? You have, safety and you have efficacy in human clinical trials, right? Anything that drug developers can do to increase the probability of success in human trials, I think is, is going to be incorporated into how drugs are developed. That's things like what we've seen, biomarkers better preclinical models, understanding different, cell to cell interactions. For example, you're trying to bring artificial intelligence and machine learning into the drug development or drug discovery process, right? All these things, aren't an effort to better predict how patients will respond to a given drug. And so that's exactly what Notable is doing.

Neil Littman (Host)
They just have a slightly different way of doing it through a functional based test, not a genetic space test, for example. Yeah, if this type of thing works in the clinic, and if it can be scaled across different indications within and outside of oncology here, I would say, absolutely, the industry will adopt this.

Danny Levine (Producer)
Well until next time.

Neil Littman (Host)
Thanks, Danny.