Tackling an R&D Project When the “R” is as Large as the “D”
Andrew Buckler is the Founder, President and CTO of Elucid, a company that is focused on characterizing placque, with the goal of providing clinicians better information on placque stability, vessel structure, treatment options, and risk factors for cardiac events. In this episode, Andrew shares how to found a company when you have a significant amount of research ahead of you, the importance of understanding biology when designing devices, the evolution of coronary angioplasty, translating research into a product, the importance of staging a project, and the problems Elucid is targeting.
Links from this episode:
Mastering Medical Device:
Episode Transcript
This transcript was generated using an automated transcription service and is minimally edited. Please forgive the mistakes contained within it.
Patrick Kothe 00:31
Welcome! Most of us in medical device work for companies that have research and development departments. However, most of the work in these r&d departments is focused more on the D or the development, than on the R, the research. Our guest today is Andrew Buckler, the founder of Elucid, a company that's tackling some really big problems. They're characterizing plaque with the goal of providing clinicians better information on plaque stability, vessel structure, treatment options, and risk factors for cardiac events. Andrew has an electrical engineering background combined with computer science. He worked as r&d Director of imaging systems at Philips, Hewlett Packard, Agilent, and Siemens for over 30 years. He was also a founding member and the first program director of the Quantitative Imaging Biomarkers Alliance. In our conversation, we discuss founding a company when you have a significant amount of research ahead of you the importance of understanding biology, when designing devices, the evolution of coronary angioplasty, translating research into a product, the importance of staging a project and the problems elucid is targeting. Here's our conversation. Andrew, you spent most of your career in the imaging segment from PET CT, MR. Ultrasound all types of imaging technologies. What drew you to that field? And why is that your passion?
Andrew Buckler 02:19
That's right Pat, I have been very interested in imaging from the earliest portion of my career, actually started Eastman Kodak that some of those modalities you talk about. There's another one you could add, I started out in digital photography at Kodak in the early 80s, where where Kodak was very proud to have developed the first megapixel CCD camera. Obviously, their interests were for consumer photography. But the majority of my professional time has been with medical imaging, I'm very drawn into the promise and not only the promise of sort of what the images look like, but what they can tell us. And those modalities that you listed. I have progressively been developed over the course of the past few decades, in order to get a more and more high quality image from a visual impression point of view, but along the way, have developed into very rich assays of the pathology that they're imaging. And so I just find that very interesting from an engineering and a scientific point of view as to how to extract the most quantitative information out of that when viewed as as a data set which has been captured by an assay and then apply the methods to it that go beyond traditional radiology and move it more into a quantitative modality.
Patrick Kothe 03:46
You received your undergraduate and one graduate degree in the states and then you extended your education at Karolinska. Tell me a little bit about that.
Andrew Buckler 03:56
It's a little bit of a it's not completely a conventional sort of trajectory. i It started out I completed a bachelor's in electrical engineering and then Master's in Computer Science. I guess I'll say this, I received a compliment. I won't indicate who the compliment was from but I and I'm not even sure it was intended as a compliment, initially, but but I took it that way. And that was that the person that it said that in his particular background, he had worked with a lot of imaging people. But he finds many of them to be remarkably in curious about what's being imaged. Now, I don't mean that as a sort of a criticism of my peers that that would be not what I'm trying to communicate. But what I would say is that this individual was sort of pointing out that he recognized that I was very curious about what is being imaged in particular that manifested itself in terms of being very interested in biology and you know, in a way that's not surprising, but in particular, I had gotten into this project and felt as though sort of an arm's length relationship. On to the biological research community would only get so far in terms of the multidisciplinary study that would allow us to reach the objectives that we have academic in sort of my approach. And so I had asked lamphead in the Department of Molecular Medicine and Surgery, Dr. All for Dean, whether he would accept me as a student, I think was taken aback by that a little bit at first because of my age. But ultimately, it has been very successful. And it's allowed me to dive deep into these multidisciplinary studies where biology and engineering can come together. And I don't believe I'm the only one that thinks that a deep understanding of multiple disciplines is necessary in order to, you know, really sort of make the breakthroughs which are scientifically possible but which are not necessarily quite hit yet. And so that's that's why that particular academic journey was part of the professional journey,
Patrick Kothe 06:02
but many of us are familiar with Karolinska, many listeners probably won't be explained a little bit about working at Karolinska is and the history behind that institution.
Andrew Buckler 06:12
The Karolinska Institute is in Stockholm, Sweden, but I think I think the way most people would know it is it is the institute that administers the Nobel Prize in Physiology and Medicine. Alfred Nobel was the Swede. And he had set up this system of these Nobel Prizes. Most of them are administered from various institutions of higher learning. And the Karolinska Institute is one of the most renowned medical universities in Europe. And arguably, it has a worldwide reputation as well. But the specific reason that I had gone to Karolinska wasn't even that I mean, that I think is necessary. More as serendipity has it, the laboratory had been collecting a very unique tissue specimen bank, that ended up being of mutual interest to the work that we were doing at elucid, as well as to support the research agenda there. And so there was a natural affinity there was a coming together of interests around this tissue bank initially, but then, you know, sort of worked out in terms of this interdisciplinary study.
Patrick Kothe 07:23
Let's dig into that a little bit. Because many of the people in the medical device field work on devices. It's something that is discrete. But what you're talking about is kind of that nexus of mechanical devices and biology, we're starting to see that more and more different groups are approaching this a little bit in a joint joint manner. Why is that? And why is it so important, as you said, to understand the biology as you're, as you're developing devices, as well,
Andrew Buckler 07:56
I'll describe it in two steps that I think folks would tend to resonate with us. The first step being devices, as you've indicated, are largely hardware, their mechanical devices, whether it be that they provide some kind of a surgical, you know, a clamp or you know, something very tangible like that, or a scanner, or a medical imaging scanner, for example. And of course, we all know, whether it be a big iron scanner, like a, like a CT, or an MRI, or, or an ultrasound scanner, they're all devices and sort of the typical meaning of that word, but but over the course of time, more and more of the actual sort of innovative content has been allocated to software components. So whether they be embedded, or whether they be interface, the device, sort of umbrella has come to to mean the software, either which controls the device and or which utilizes or exploits the information produced by the device. And so, for example, the FDA manages its regulatory oversight of software within largely the Center for Devices. And so this is this is a device. But the second step then would become the question is you acknowledge it is sort of what would this software do? Well, in medicine, of course, what we're trying to do is we're trying to make a contribution to patient care, what is patient care, but some biological process, which is underway for good or for bad, and of course, oftentimes, the patients are having it for bad and so it's understood as pathogenesis. And so if we're to help with that, then we should understand what that pathogenesis is. And then moreover, to the extent that software might be evaluating whether it be an image or whether it be some other signal, one idea would be to stop early in terms of that understanding and have the clinicians take it further But increasingly, the developers of this device software, myself included, want to and are sort of driven to provide more value. And so whether that be for example, in the area of precision medicine, so that the measurements that we take the images we take, etcetera, can contribute to the optimization of therapies for for patients, you know, or for, you know, related purposes, you know, diagnostics or therapeutics, it all sort of comes down to the fact that understanding that biology so that engineered solutions that can first elucidate it, and second do something about it, once it has been elucidated, are increasingly sort of in the vanguard of what device software would, would do, I'll just briefly mentioned, there is sort of a sister field, and I think many of my colleagues are participating in in this field as well. And that is sort of biomechanics or sort of physics based modeling, I share their enthusiasm for those techniques as being relevant. However, the the biology, focus is, is capturing not only physical interactions, but But in particular, those interactions with a very, very high number of degrees of freedom, what we call life, there's a utility, but sort of a glass ceiling in terms of what you can do with biomechanical modeling. As a result of the, you know, the assumptions you have to make until and unless you sort of confront the biology itself, which, which has sort of a richer set of terms within which to describe living systems. There is a relationship between what can be done with biomechanical physical models, but we go beyond that in terms of understanding biology and a more dynamic, higher degree of freedom system.
Patrick Kothe 11:56
For this discussion, I mean, there's there's combination products that have devices and biology or pharmaceuticals associated with it. For this discussion, we're really not talking about that type of thing. We're talking about understanding the biology, as we're, as we are putting an interventional product out there or a device out there to intercede.
Andrew Buckler 12:18
Yeah, I think that's fair to say. I mean, if we were to think about, for example, molecular diagnostics, that that's a product category, a set of devices, typically, they are performing some kind of an ohmic analysis on a sample volume that, you know, what we're doing is not, it's not dissimilar from that, if one considers the sample volume to be an imaging data set, which may be represented as a three dimensional volume of data, it's not dissimilar from whether it be a blood sample or some tissue which has been biopsied. And then the question is sort of what is it that that sample? What is it telling you about what is occurring within that sample? And then are looking forward in terms of well, okay, on the basis of this, what might come next? You know, what could we do maybe to improve? What comes next? You know, part of that might be to understand how did we get here? What were the drivers from a molecular or cellular level that sort of accumulated into this particular disease presentation, insofar as they can help us understand the present, and or improve the outcome as as it goes forward. So in that sense, it's a little bit different from a therapeutic as such, there are related fields. So for example, theranostics is sort of a therapeutic with imaging components sort of together, manifestations of our technology might be, might be utilized in real time during an interventional procedure, for example, maybe with a CRM, CT scanner to be to be looking at, at tissues directly. And in the course of a specific intervention. There's also the idea of both identifying in advance which procedure will intervention might be most efficacious, what the predisposition is for a negative effect for that, whether it be a direct adverse event or for example, restenosis and a stent, or for taking a product, which is a combined drug and devise for example, a drug eluting stent, and making a selection of which drug is it that would be the most effective with respect to the the manifest phenotype of the host of the patient that we're looking at. So there really are blurry lines between these things. But the bottom line for us is that we're taking a sample volume, in our case the image and we're determining from that what is going on at levels that are not visually conspicuous
Patrick Kothe 15:01
when you look at just the location of of a, an Oregon or the target that you're going after, and as you're doing the procedures, you're, you're checking the location of that organ. So that's one piece. The piece that we're going to talk about with lucid today is beyond that. It's the characterization and biology of that particular structure that's ends, right. So, I want to tee tee this up with my experience in the early days of angioplasty. So the early days of angioplasty in the coronaries, the main concept was to put a balloon in an artery and just smoosh back whatever's there, into into the wall and create a channel. That's right, and to create that checks, right? That was that was kind of the early days and, and some of them stayed open. Yep, some of them do. And, and then as they didn't, people started saying, Okay, I'm going to attack this from a device standpoint, let me see if I can shave some of that off and use an atherectomy devices to pull it out, right. And then there were other devices to say, I'm going to spin something really, really fast and emulsify whatever is in there, and I'm going to create that channel, as well. And some of that worked. And some Sometimes it worked, and sometimes it didn't, then there was a company called Seavus came out with with an ultrasound device, let me let me start to see if I can characterize some of this stuff. So some work kind of went went out in that direction. And then you know, we we know stents hit and drug eluting stents, it because we started to understand that the tissue and the understanding of the biology in the coronary artery was important to making an interventional decision to try so that's TNS up, or what what you guys are doing a way to think about, here I am as a device, and I'm going to fix something without understanding of the biology. And now we're going to understand it more so we can have more precision medicine.
Andrew Buckler 17:07
Yeah, I think that's very well said. And I think what's evident in sort of that, I think, accurate, historical run up, is that it begins with measures of structural anatomy, relatively simple things, is the lumen paid? And how big is it? You know, can we make it bigger, and there's, there's a geometric concept going on there. From a diagnostic point of view, from an imaging point of view, you'd want to see it, you know, maybe you'd have a planning process of some kind of determine, you know, how big the device should be tactical aspects of how to get that device there. Then that tissue that was in there that was that was that was causing, in your examples, flow limitations? In what is the nature of the tissue? Is it? Is it something that you can expect to sort of push out of the way and remain pushed out of the way? Rather, is it something that has a pathogenic processes just going to reach the nose? You know, is it the kind of thing that if you cut that out, that it won't come back? You know, that it itself is the bad thing, but that it's there's not a generator of that bad thing? That's just gonna follow you right, right afterwards. And then all of those procedures that you mentioned, are all late stage disease. I mean, I think folks that are sort of listening to this, certainly myself and my loved ones, I hope don't have to have those devices, as helpful as they are and as positive as they can be. That it somehow would seem better somehow to not get there in the first place. You know, what can we do with pharmaceuticals? For those patients that can't necessarily make lifestyle lifestyle changes? You know, what can we do with pharmaceuticals, and there's a certain bluntness to those as well, initially, right. So, you know, the first blockbuster medicine, I guess, was aspirin. It's it's a wonder drug. You know, one wouldn't describe it as a specifically targeted drug, not an unlike your sort of rendition of the history of coronary interventional devices. You know, drugs have gone through from very general to much more specific. So whether it be procedural interventions, or whether it be pharmaceuticals, it comes down to a similar need, at the present time and in our lives. And that is to be more specific, so that the selection of what to do can be based on the underlying cause, as opposed to doing something which whether it be having a temporary effect or maybe even no effect at all in the worst case, and then just come back and we haven't helped the patient. All we've done is to burden society with the cost of doing an intervention that actually didn't help.
Patrick Kothe 19:52
Going back to that understanding to in the early days. I don't know if it is well understood what the needs If the blockage was and what I mean by that is think it was understood that there is an artery and that artery was like a pipe, and you've got stuff flowing inside that pipe that blocks that flow. And over time, we've discovered, it's not only stuff that's inside that pipe, but the walls of that pipe are affected. And that kind of changes the game a little bit in terms of how you treat that, how you how you're going to have an interventional product to treat.
Andrew Buckler 20:32
Right, I can pick up on that by saying that the first daughter understanding of vessels, of course, is that they're conduits. The most important part of a conduit is is what is its inside boundary. Of course, in vascular, we describe it as the lumen. And so, you know, understanding the lumen, being able to know what the shape of the lumen is being able to know whether that lumen is encroached on by something inside of it, which is, you know, causing it to be narrower than it otherwise would be, you know, are all very natural for store to things to understand, and there's been a lot of study, this is a lot of devices that sort of work at that level, but even recent devices still work at that level. And they can, they can have a contribution to make. But if we understand that the reason a lumen may or may not be problematic, is because of the wall that surrounds the lumen that that whether it be that there's some kind of a tissue that is has encroached the lumen, or whether it be you know, tissues that can sort of break off from from the wall and make their way into the lumen. All of these things originate in the wall. And of course, groundbreaking studies from you know, along, you know, many, many years ago, have discussed the nature of remodeling. For example, there was there was work in the 80s by Glenn Goff, which talked about the walls remodel outward. First, that there's a compensatory biological mechanism that which sort of preserves the lumen. However, at some point, that remodeling overwhelms that compensatory mechanism and starts to remodel inward. And of course, we describe that as stenosis. We also know that, you know, through through the decades of study that lipids, lipid species, which are in the blood, can make their way into the wall. This can happen anywhere. But there there are sites that which is that it's predisposed for this on the basis of hemodynamic behavior. In any case, once those lipids are, where they shouldn't be, instead of in the in the blood in the wall, they can start to create cascades of of pathogenesis. More recently, we've also learned that what amounts to a very similar effect can can be triggered actually not from the wall out into into the lumen, you know, sort of an inside out hypothesis, which remains the dominant hypothesis around atherogenesis. But that systemic inflammatory insults from other, which manifests as other disease conditions, for example, psoriasis, rheumatoid arthritis and other diseases like this can also cause effects inside the vessel walls, irrespective of serum lipids. It's been described as an outside in mechanism, you know, whether it be the one mechanism or the other mechanism. It's only through understanding what's happening in the wall that we can project, what would happen in the lumen, maybe some level of it is already happening, maybe it's going to happen a little bit more. And so you can't understand the lumen without understanding the wall. It's an overused adage, if if adage is the correct word for it, where there's this man looking for his keys in the night, and there's a policeman who's asking him, you know, what he's doing? He says, I'm looking for my keys. And the policeman said, would, you know, he's underneath a streetlamp. And so, you know, where did you lose your keys? Oh, I, you know, left them over there. I mean, I lost them over there. And, you know, why are you looking here? He says, Well, this is where the light is. I mean, so I'm going to look here, but, you know, it's easier to study the lumen than the wall. And so, you know, we've done that as as a as a society, that's important, but it's not going to get us further. We now need to go back and look at what causes not just what the effect is.
Patrick Kothe 24:43
So you have been on industry side and you are a maker of devices. There's another side and it's the research side, making a product out of research is kind of a difficult thing. So let's talk a little bit about a risk Search project versus a product development project, you very eloquently talked about the the idea of understanding the biology. But how do you get from understanding to having a product? Because sometimes understanding is just good enough to develop product. Right,
Andrew Buckler 25:20
right. So, you know, there's obviously a lot that can be said on that. And it's highly situational dependent. But I guess I'll tee it up this way. In some degree, what we're doing isn't new. And what I mean by that is that it's been recognized for a long, long time, we're not the first ones to recognize it that luminal stenosis, despite the fact that it continues to be the dominant decision maker for clinical care guidelines, is insufficient. And there's been research for decades, into fact biology, we know that plaque biology is not only causing the stenosis, or more recently, the focus on low reserve deficits, that's not new. What's new is to translate these understandings and is something that can be used practically, in clinical guidelines. And there's a chicken and egg dependency between the availability of devices. And the guidelines, which both prescribe the use of those devices, as well as identify decision thresholds for how to interpret their output. I, for example, would say that our particular field has been hampered by a, shall we say, a less appetite than I would prefer of the research community to sort of move what it's doing into practice, through commercialize available devices. And by that I don't mean to criticize or to suggest that the research shouldn't go on, but rather that there's a sort of a rich vein of work that that will will will impact patients to the extent that we can translate this understanding and get it, get it into a device and then get that device into the hands of clinicians so that they can use it for their patients. So the question then becomes how well there's all manner of obstacles. So one, one obstacle, we've already identified this, this cyclic dependency between the existence within the clinical guidelines of the use of such devices, but of course, those guidelines are based on on data and evidence of the efficacious or lack thereof of a given approach. But if that approach isn't available, you can't collect the data on which to determine how effective it is or isn't. So so the first problem is somehow to crack the nut of the of that cyclic dependency, I think that then parlays into the idea that you have to be developing devices with some level of investment, fortitude, the value of this is worth the time and cost it takes to develop it. And for there to be forward looking means of of investing this this commercialization activity, and grants play a big part of it, you know, the right investment vehicles play another part of that. But then there's the question of sort of discipline and repeatability of science, you know, many experimenters, you know, not taking anything away from them as such, in terms of sort of the individual contributions they make, but I think it's broadly recognized that the ability to replicate those results and scale them up is itself difficult, you know, even even in irrespective of the specific thing which which might or might not have been discovered at research. And so there's all manner of real world problems to deal with there, for example, repeatability and variability are words that are not always used in research circles, but which are the grit and grist of being able to do a robust scale up. But those themselves have scientific underpinnings in order to properly understand them and and to sort of bring them under control. And that takes hard work. And that that work is in and of itself, maybe not the most glamorous portion of it, but it ends up being some of the most important in terms of translating from a research effort into into a commercialized effort. But then last but not least, I'll just say that there is an innovative component it's not you know, that we've already known all this from decades ago. The part of it that we haven't known is the connection. And so in our particular case, what I would say is that to the extent we have for example, a CT angiography image. We know the types of mechanisms that are occurring in the population at a biological level. We also know the type of presentations that we find when we image our patients. What we haven't been able to know and where the innovative portion for us comes in, for example, and by analogy, other other groups like ours, is the connection between those two. So at an individual level, it's all very well and good to know, you know, about atherosclerosis as sort of an academic study in the population. It's all very well and good to make images of individual patients. But how do they come together? What's the biology for this patient right now? Not as a general issue, but as a specific issue, because because it's only in that level of specificity, then that we get to Okay, on that basis, this therapeutic direction is appropriate for you know, patient a, as opposed to this other one for patient B and having some level of confidence around that.
Patrick Kothe 30:44
So let's talk about a lucid how you started, the problem that you're trying to solve. And the mechanism that you're using
Andrew Buckler 30:52
the lucid sort of impetus came from both an opportunity and a need. And, you know, my sense, initially, as founder of sort of trying to match the opportunity and the need, what is the opportunity? Well, the opportunity is, as I was indicating, in some of the introductory remarks, these images have become exquisite, visually, they're impactful, you know, it's a modern miracle of medicine to be able to see into the into the human body, the level of detail has gone up and up and up. Now, this detail is manifest in part by spatial resolution, but in part, by contrast, resolution, and with different types of scan protocols and a variety of ways to to improve this as an assay, whether it be through approaches to contrast, administration, whether it be different modalities and their relative strengths and weaknesses, we've got this very rich assay. And what it does is it takes non invasively tissue and in essence puts it in your hand. You know, there's many sorts of ways this can be brought to bear a variety of problems, one of the largest areas of involvement of the use of what's called quantitative imaging, if you will, is in cancer. Oncology has had the benefit of high level funding. It's obviously a disease area people care about. And there's been great progress in oncology across the board, but But in particular, about quantitative imaging, in oncology. But my observation is that is as important as oncology is that cardiovascular disease remains the single largest cause of death and disability, and not just in the developed world, as the developing world becomes more affluent, and maybe takes on some of the downsides of the developed world in terms of lifestyle and diet and things of that nature. Cardiovascular disease has been rising up, but different than oncology. You can't buy up see cardiovascular tissues for the most part. Now, there are ways to get tissue, but you don't generally biopsy you can't like poke the plaque, grab a little piece of tissue and figure out what it is, you can actually create the very thing that you're trying to avoid by doing that. It's sort of you know, it's just sort of a non starter. And so the question then becomes, well, okay, maybe then you can't have any information about the tissue, maybe all you can do, the best you can do is blood tests. Well, blood tests might or might not be okay. It's not for lack of trying that there is found a difficult, you know, various investigators have worked on this with a lot of a lot of spended. This is a big area, of course, but there aren't specific markers. Even if they can get the specificity up under certain conditions, you still don't have the location. So particularly for an interventional procedural intervention, you need to know the location. But more than that, even for pharmacotherapy, you need to know the distribution of disease, because that tells you something about the pathogenesis. So bring the opportunity of the of the quantitative imaging methods, together with the problem of cardiovascular disease burden in society, try to make a contribution there. The term precision medicine has received a lot of attention, mostly in the cancer domain. Can we do that for atherosclerosis and cardiovascular disease as well? So this is sort of where these things come together for lucid. And then the last thing I'll say just sort of about the formation of a lucid is, as you were, I think, correctly identifying this research and commercialization, you have to have a business model that would allow these to come together. And that's not obvious. It's certainly the case that there's an established business model in pharmaceuticals, which allows clinical trials and study to have the time and resources that they need But in other disease or other product categories, specifically devices, that tends not to be as as large a budget or as long a time available for clinical study. That doesn't mean there's no study. Of course, that's absurd. Of course there is. But in order to really go deep into mechanism of action, and pairing those two together, and determining a software analytic that can help in that requires a very deep study. So I've often described that there's a continuum of business models and the quantitative imaging biomarker development, which is what in essence, we're doing the category, if you will, requires an intermediate one, we need to have enough time and resource in order to incorporate within the software analytics, what is the most salient points of the biology and how to solve it. To do that, what we need to do is we need to develop the ability to have whether it be pricing models or you know, time in the marketplace, or other ways to monetize that, in such a way as to return an attractive return to investment capital market to the equity markets. And so that's sort of the other piece. So it's not only about the technology and the science, that elucid, which are some of the things I have been discussing, but as well marrying that together with a business model that can allow, you know, that can be attractive for investment capital.
Patrick Kothe 36:30
Typically, within a medical device field, it's small, are largely sent some, some research but mostly developed. That's right. And most of the research takes place in the university setting. What when you're trying to tackle a large problem like this, you've got a lot more research to do before you get to get to develop and development and whether you are ever going to get to the vets, right, it may strictly be a research thing. So how did you guys approach that you knew you knew you had a big problem here, you knew you had I had a lot of research to do. And I assume you had to have a pretty complex research plan to be able to assess different timeframes, how much money you're going to need, whether you're going to go into the VC community. Okay, how did you approach that problem?
Andrew Buckler 37:20
I'm glad you're asking, I'll start by saying my father was, was was a semiconductor physicist at at Bell Laboratories. I think those of us that are old enough to sort of recall that what that was, was that was a big R and a big D, organization and provided clear and distinct benefits economically and in many ways. And so I this is what I grew up in, I just was sort of steeped in this idea of our being a full participant in r&d. The tactical aspects of what you say is that one ideas that you can't even get started because no one will find you because what you're working on is too far out. And so for us, what it's meant are two aspects that I can highlight relatively simply, one is that the scientific nature of what we're doing is multiscale. That's an area of analytics, you know, it's sort of a descriptor of, of a form of the analytics that we do. And so you can proceed stepwise. So for example, in our case, we proceeded to tissue characterization at a microscopic scale with histology as ground truth first. In other words, irrespective of protein to protein interactions, can we first identify the difference between for example lipid rich, necrotic core and intraplaque hemorrhage? That's a hard problem already. But it's not as hard if you will, is going the next level then to get down to protein to protein interactions. Staging it out in steps is sort of one practical way, obviously to to be able to take a complex problem and break it down. But then that is necessary, but not sufficient to make a sustainable business model that can run sort of all the way through it. With regard to sort of the commercial revenue capacity and commercial traction, then it becomes, you know, what can you productize early that can provide the impetus to get the next round of investment on the one hand, and provide some level of commercial traction on the other so that people can see that this is not just an ivory tower research project. So for us, we started with plaque morphology characterization. So elucid is very proud to be the first and in fact the only that that has FDA labeling that's specific to tissue types. For example, we don't do only sort of a broader term called low attenuation plaque we have a specific labeling for lipid rich necrotic core which which has a histological definition, but that could be sold it is not a high volume, you know there's not a lot Have uptake in terms of clinical guidelines yet, but it is saleable. And so plaque morphology is the first step and in the sales sort of roadmap. A second one is the the development of diagnostic phenotypes. And for us, we have two, a one of them is histology, defined high risk plaque, which is associated with the propensity of the plaque to rupture. And the other is FFR, CT fractional flow reserve, as measured by CT, it's that latter one that has that enjoys the greatest level of reimbursement right now. ffrct, there we are, a fast follower to another call a competitor or colleague company, which has done a CTA based FFR. We do that as well, we do feel that our measurement of it is more direct than the other method. We believe that we have various advantages over the other method, but we congratulate you know, the other company for having put an ffrct product on the market. And with that, that there's the establishment of reimbursement codes, etc. So they the revenue can be robust in this intermediate period. The other technology can't necessarily extend though to our next step in the roadmap, which is the tailored therapeutics, not only do we identify the risk of a given for outcome, but we, you know, start to elucidate and be able to select among the therapies, the revenue and the and the sort of the investment attention that we provide in the mid part of our roadmap then allows us to extend to the latter part. That latter part is very, very interesting for many investors more interesting than these earlier steps because it would eclipse those other steps in terms of overall business opportunity and potential. But they're admittedly longer range studies to to complete that development and validation. This strategy of our sort of three step processes our approach to that.
Patrick Kothe 42:02
So your product is what hardware hardware product software product does it bolt on to another imaging technology, tell me about your product,
Andrew Buckler 42:11
we have three product deployments. So the product itself is to create these measurements, classify the phenotypes predict outcomes and select therapies that is a collection of functionality that could be broadly described as a clinical decision support system. We deploy that in three different types of deployments. One is we can conduct the analysis here, in other words, establish a portal with the imaging and or the clinical covariates from PACs and EMR. We conduct the analysis and we give back a report for individual patients. We can also and we do conduct research studies of cohorts of patients and we do a variety of forms of statistical analysis that the cohort level and that's a saleable sort of catalog item. And what is the imaging source? So CTA is what we're using now we've done working in other modalities we have worked in MRI as well, but but CTA is the primary source, CT angiography, CT angiography. The second, though, is that we can also and there are some companies in our space that that do the conducted analyses, but we're the only ones that can also deploy this software onto an individual site. There are many sites, whether by policy or by law, cannot share the information outside of the out of the organization's boundaries. We've taken the time to develop the user interface, the ability to use this by people on site as well. So that is another way of doing it, rather than are conducting those analyses on an individual patient basis. This breaks down into two we've got a client server relationship, or excuse me architecture, we can deploy just the client, but use our server as a cloud resource. Or we can deploy both the client and the server together on the site. So these are the three deployments sort of manifestations of that product.
Patrick Kothe 44:17
So the product itself is your it's an analysis engine, it's a software, software engine, you're using CTA data coming in, you're analyzing. So there's different ways that can be used. We can we can identify issues, we can diagnose issues, we can track over time, what's going on with somebody as we're aging, or we can use it as immediate during an intervention and then follow up post intervention, correct. vention where's your sweet spot?
Andrew Buckler 44:48
Yeah, we start by guiding the selection at relatively end stages of the disease. So whether patients are directly symptomatic or if they're asymptomatic, but there or assigns or other risk factors which could could cause them to be analyzed. What we do is we make good decisions about the utilization of procedural intervention. So that's the sweet spot right now, that's where the reimbursement is. Measurement of ffrct is well accepted. And adoption is going up, particularly as CTA itself is going through a parallel increase in terms of how well it is represented in terms of recommendations by by the by the medical community. And so the sweet spot there is to make good use of that cath lab to get the right people in there that are going to benefit and those that are not going to benefit not to bear the expense of a diagnostic Cath, which is an unprofitable use of the cath lab and clogs up the resource for those that could use it better. And then, you know, the beginning of the pharmacotherapy value proposition starts by if there's not an urgent need to get into that Cath Lab, there's a window for pharmacotherapy, let's use it. Let's see if that was effective or not. Okay? If it's not, then they go to the cath lab. But but the point is that the utilization of the cath lab is for actual needed procedures, as opposed to wasted and diagnostic cats. And then the natural extension is what we hope is that the sweet spot then then moves up in the disease process to primarily being pharmacotherapy, and what we do is for pharmacotherapy is we do a similar value proposition to the cath lab. And that is to make better utilization of intensive medical therapy, for example, PCs, canines, anti inflammatories and other emerging disease, drug categories, which are very expensive on the one hand, and or which have side effects. And so very similar to making good choices about who goes to cath, making good choices of who gets a PCS, K nine, for example, versus anti inflammatory things like that. So sweet spot today is to make the best use possible of that Cath Lab, sweet spot as we move forward is to move that up and make good choices about how to best utilize pharmacotherapy,
Patrick Kothe 47:16
as you're moving up in up in the line, or I guess back back in the line, younger and younger patients and making predictions on where they're going to end up with an intervention, I imagine there's some machine learning some AI that you're employing in there as well, there is,
Andrew Buckler 47:33
what I will say is that we're not among those companies that sort of consider AI to be, as it were, the central focus for us the central focus is what this application is, what I mean to communicate by that is that AI is not a panacea. It's not something to sort of be taken, as it were naively. Because you know, as many strengths as it may have, those strengths are best utilized if one understands his weaknesses, and not just understands them, but in fact understands how to mitigate and sort of position it. So we we treat our classic image processing people with with equal dignity of our of our deep learning engineers, for example. And in particular, with the deep learning that we do, we practice a form of modeling that's referred to as interpretable models. There's a continuum between black box and ultimately interpretable models. Explainable AI is a little bit of a current buzzword in between that, where one takes a model and one then evaluates it afterwards to try to identify what it's picking up on, that has certain biases, we see what we like to see what's better. And the theoretical construct of interpretability articulates a vision where you know, what the model is developed for in the first place. It's architected in order to do that. And the resulting information is interpretable. An example is with our ffrct that we've done is other methods have what amounts to a mathematical model, machine learning or otherwise. But it's very difficult to say on a patient by patient basis, why it's coming up with the number that it's coming up with. In our case, we start with the cause of flow deficit, which is the vaso, dilatory capacity the walls, we visualize and quantify that vaso dilatory capacity, that is to say the drivers have that capacity. And with that the fractional flow reserve measurement is easily interpreted by clinicians visually and quantitatively, without, you know, sort of going into this idea of weighing. How did they come up with that number? It's right there and you can see the biological rationale for it. So all to say that, that we are among those that we consider to be and we hope not in modestly, to be the most skilled in the application of deep learning and other forms of artificial intelligence to this problem area, but not because that's all we do, or because we do it mindlessly, but rather deep insights, specifically, with the theoretical underpinning of model interpretability, at every step of what we do,
Patrick Kothe 50:21
you mentioned that some possibilities in your business model is to ship you the information and you provide the information. And that may be good for the early stages. But in the acute phase, where someone's got somebody on the table, and you're doing it, they need immediate information, right? So from a business model standpoint, how do you monetize that type of of a product, when it's utilized with CT going at the same time,
Andrew Buckler 50:48
we have two sort of approaches to that one, and it depends upon the contract with with the using institution. One is there, there are estimates of volume of how many patients that will be used on how, you know, bigger the data sets is other volume measures that we use in order to set the licensing cost. There's also the opportunity to have a per use fee. You know, that's that's sort of another way to be able to give fair pricing in terms of how much they're utilizing the technology.
Patrick Kothe 51:24
So you have a big problem with a large r and a large D with it. How long has the company been going?
Andrew Buckler 51:32
I started this not quite 14 years ago. This is a good news, bad news story. I you know, some people will say, you know, oh, my gosh, you know, that's too long. And what I would say is that, listening to the other aspects of this discussion, and comparing it with other other efforts that have made made big, our efforts, when I talked to those people, they're not surprised as Oh, good, that actually instills confidence, because, you know, there's the idea of there's two positives there. One is you have been proceeding step wise, you are going through a methodical, stepwise process to develop the right to the right develop the right base. And there's no better way to do it than to do that. But as well, and it demonstrates staying power. I mean, you know, it's hard to stay in business, you know, for a long period of time without making a contribution. We've also enjoyed, I think, a very successful application of grants, I'm not necessarily a big government kind of person. But what I will say is that the notion of the government, being helpful in the development of these kinds of technologies, by having different grant programs is something certainly which has benefited us with and we believe, in turn will benefit patients, we've received healthy grant funding from the National Institute of Standards technology, the National Science Foundation, and the National Institutes of Health. Without that we would not be here, I mean, that that's, that's candidly clear, in recent years, we have turned our attention away from grants to instead, you know, the more typical investment vehicles, there are some companies that sort of get stuck in sort of a grant zone, you know, and it's as if they exist, just to get grants, I, I have never been a huge fan of that. What I'd like to do is I'd like to take advantage of the grant programs to the extent that we can make compelling applications, use that for that early stage and then transition into the investment capital and ultimately out of the investment capital and into, you know, a normal company with solid sustainable revenues, you know, that. So in that sense, I'm a little bit of an old school, kind of a guy. But but that's that's the track run.
Patrick Kothe 53:40
How far away are you from that sustainable revenue?
Andrew Buckler 53:43
It depends. So So this falls announcements we have, we have a couple of very exciting applications into into the into the FDA right now. You know, I don't want to sort of priests, priests, suppose what the timing of those are going to be, but they should fuel what could be a sustainable business, just on the basis of sort of the ffrct level, I mean, that can easily create a sustainable revenue model. We we're not stopping there, though, as we've been talking about this area of tailored therapeutics, that will require investment for another another few years. But what we hope is that with that, there would be, you know, an even bigger prize at the end of that, and then based on, you know, how effectively we're able to execute on that plan, then it would be a few years out, but it would be a few years out on a bigger prize. So it's you know, two different ways of looking at it. You're
Patrick Kothe 54:41
not on the market yet.
Andrew Buckler 54:42
We do have no we do have product on the market. We don't brag about our revenue. We don't you know, say that you know we're killing the world with, you know, with what we've got right now, but no, we have we're an active selling we have two configurations. We have a configuration for clinical use what it does is it measures, plaque morphology. So the structural measures like stenosis, remodeling ratio, that kind of thing, but also lipid rich necrotic, core dense calcification with the best estimates that exist on the market with the only labeled FDA labeled performance metrics of their accuracy. No competitor has that other than us. And then as well, we have a research configuration, which our strategy has always been to have a research configuration, which has the capability that is presently in regulatory, but now already available for research use. And so that configuration has, has the histology to find high risk plaque phenotype as well as the ffrct already. And so that's also saleable. These fall regulatory applications within take those two and put those into the clinical product as well, in the space, we hope of very few months from now. And what was the regulatory pathway, this started you define HRP as Lenovo, Lenovo 510 K, the ffrct is a straight straight out 510 K, the the other company that I mentioned, who created the first ffrct Establish establishes a predicate device. So I don't want to say anything. And regulatory is easy that that's not the case. And if you have anyone on your podcasts that were to say such a thing, then boot them off. But but, you know, I would say that it's manageable. And we have a high level of confidence not so not so difficult. The area that we have for the Lenovo, it's new in two ways. One is it is what had not that many years ago been described as a Holy Grail of modern medicine to detect the vulnerable plaque, non invasively. So it's novel in that respect. But it is also novel in its methodology. What it does is it uses a category we know computer aided detection, we know computer aided diagnostics, this is referred to as computer aided phenotyping. And there's certain study design, the nature of the validation data and the like would be would be new regulation. So it's a de novo 510 K. Going forward. We're assessing and we're in communication with with FDA and other regulators around the world on the most appropriate pathways for the tailored therapeutics.
Patrick Kothe 57:24
Andrew, this has been an excellent discussion, I really appreciate what you're doing. Many of us are going to be utilizing technology like this not only for tracking ourselves as as our aging, but also we're going to end up having interventions as as well. So thank you for the work that you've done on the research side of things. And now on the development side, it's fantastic for those people that are attempting to solve big problems, like you have done, what kind of words of advice would you have for him?
Andrew Buckler 57:57
This many sort of philosophies around this idea of grit? Yes, maybe it's almost immodest to sort of apply the word grit to yourself, I, what I would do that was I would, I would say, you know, buckle up, you know, don't don't expect sort of a quick thing. I mean, you know, do the hard work, don't necessarily try to skip steps, because that won't save you time, you won't get there and you know, do the due diligence. I'll also say that we sort of presuppose, it's a big problem, there are many people that work on it, that may wish to work on sort of an incremental sort of sort of refinement of something, if you're going to put you know, whether it be 14 years, or fill in the blank number of years into something, you can't do that many times in your career. So, you know, skip the sort of the easy refinements, you know, skip the things that only move it by, you know, five 10%, you know, go for the thing that's going to make a wholesale change. Now, you may fail, and that's okay. I mean, you know, there's a lot of people that describe failure as a positive outcome. I don't want to like candy coated and say that failure is great. But what I will say is that, if you're working on something that that is substantial, and that would, would really move the needle, it's worth the time, think it through, you know, sit in the chair, or whatever the analogy is, and just, you know, do that work. If this story is a good one. There are certainly other stories as well of entrepreneurs who who have done that and have made a positive contribution to society that was worth the time.
Patrick Kothe 59:28
So many of us are focused on a development, sales and operations. Thank goodness we have people like Andrew and companies who are devoted to understanding and tackling some of these huge, long lasting projects. A few of my takeaways, Andrew discussed, his dad being involved with Bell Laboratories in the vision of Bell Laboratories, and that really kind of describes this big our issue. Bell Labs arteries really was focused on solving these big problems. They encourage their employees to go deep into technology. And part of their mission is solving the challenges of human needs before they exist. These are some of the really big problems that people are focusing focusing on. And that's what Andrews worldview of research and development was really shaped by by his father being part of Bell Laboratories and, and as he said, are being a full participant in r&d. The second thing was a lucid has been focusing on this for 14 years. And many of us think that that's a really long period of time. But when you're, when you're trying to solve these big problems, it really is necessary to spend a good amount of time on it. And when you think about this, if you are devoting yourself into a company like this, and this is obviously a significant portion of your career, make sure that the vision is worthy of your time investment. The last thing is, as he described handling a huge project, yeah, how do you do it, and this is something that that he spent a bit of time on was proceed step wise, stage your project out into steps and and kind of tackle these steps. And what he said was, this is necessary but not sufficient, because you can't really fund a company this you this way, you have to have some early commercialization. So getting getting out step wise and getting some product out to the marketplace is really an important part of that. But then, once you get that, that incremental product out there, you have to focus on that home loan application. Thank you for listening. Make sure you get episodes downloaded to your device automatically by liking or subscribing to the mastering medical device podcast wherever you get your podcasts. Also, please spread the word and tell a friend or two to listen to the mastering medical advice podcast as interviews like today's can help you become a more effective medical device leader. Work hard. Be kind
