lystn

If you’re a parent, you’ve probably had this experience many times: Your young child has a high fever, and maybe a sore throat, but you don’t know exactly what’s wrong. Is it a bacterial infection, in which case an antibiotic might help? Or is it a viral infection, in which case, you just have to wait it out? The symptoms of bacterial and viral infections are often the same, and most of the time, even a doctor can’t tell the difference. Viral infections are more common, but sometimes, the doctor will prescribe an antibiotic anyway, if only to help the parents feel like they’re doing something to help. But what if doctors didn’t have to guess anymore? What if there were a fast, easy blood test that a doctor could run in their own office to look for biomarkers that discriminate between bacterial and viral infections? Well, that’s the seemingly simple problem that a company called MeMed has been working on solving for 13 years now. Recently MeMed’s first testing product got approval from the FDA, and now the company is finally beginning to roll out it out commercially in the US. And here today to tell us more about how it got built, how artificial intelligence fits into this picture, and how rapid diagnosis could change the practice of medicine, is MeMed’s co-founder and CEO, Eran Eden.

Please rate and review The Harry Glorikian Show on Apple Podcasts! Here's how to do that from an iPhone, iPad, or iPod touch:

1. Open the Podcasts app on your iPhone, iPad, or Mac. 

2. Navigate to The Harry Glorikian Show podcast. You can find it by searching for it or selecting it from your library. Just note that you'll have to go to the series page which shows all the episodes, not just the page for a single episode.

3. Scroll down to find the subhead titled "Ratings & Reviews."

4. Under one of the highlighted reviews, select "Write a Review."

5. Next, select a star rating at the top — you have the option of choosing between one and five stars. 

6. Using the text box at the top, write a title for your review. Then, in the lower text box, write your review. Your review can be up to 300 words long.

7. Once you've finished, select "Send" or "Save" in the top-right corner. 

8. If you've never left a podcast review before, enter a nickname. Your nickname will be displayed next to any reviews you leave from here on out. 

9. After selecting a nickname, tap OK. Your review may not be immediately visible.

That's it! Thanks so much.

Transcript

Harry Glorikian: Hello. I’m Harry Glorikian, and this is The Harry Glorikian Show, where we explore how technology is changing everything we know about healthcare.

If you’re a parent, you’ve probably had this experience many times: Your young child has a high fever, and maybe a sore throat, but you don’t know exactly what’s wrong. 

Is it a bacterial infection, in which case an antibiotic might help?

Or is it a viral infection, in which case, you just have to wait it out?

The symptoms of bacterial and viral infections are often the same, and most of the time, even a doctor can’t tell the difference.

Viral infections are more common, but sometimes, the doctor will prescribe an antibiotic anyway, if only to help the parents feel like they’re doing something to help.

But what if doctors didn’t have to guess anymore? 

What if there were a fast, easy blood test that a doctor could run in their own office to look for biomarkers that discriminate between bacterial and viral infections?

Well, that’s the seemingly simple problem that a company called MeMed has been working on solving for 13 years now. 

Recently MeMed’s first testing product got approval from the FDA, and now the company is finally beginning to roll out it out commercially in the US.

And here today to tell us more about how it got built, how artificial intelligence fits into this picture, and how rapid diagnosis could change the practice of medicine, is MeMed’s co-founder and CEO, Eran Eden.

MeMed has a growing office in Boston, but I reached him at the company’s first office in Haifa, Israel.

Harry Glorikian: Eran, welcome to the show.

Eran Eden: Thank you very much for having me.

Harry Glorikian: It's great to have you here, I know that there's a significant time difference, so I appreciate like but it still looks like it's really bright and shiny out there right now. So what time is it in in Israel right now?

Eran Eden: Five o'clock in the evening,

Harry Glorikian: It's five o'clock. All right. Well, you guys have a lot more sun than we do anyway because we're in the middle of winter, but absolutely.

Eran Eden: So this, here, is actually full of people as well. So yeah, you don't stop innovation as five o'clock in the evening.

Harry Glorikian: So, you know, I was looking at your background and I mean, it's really it's interesting. It's diverse. You have a degree in biology, computer science, systems biology. You were first job was in computer vision data and analysis. But then all of a sudden you wound up starting a company that builds sensors and software for infectious disease. Like, how did you end up down this path, and do you feel like everything that you were doing until you got here was preparing you for it?

Eran Eden: Well, I think... A great question. So I think, on the face of it, it obviously the background in data science, as you know, in molecular biology, obviously all of that relates to what we're doing is part of our day to day and it is a good starting point. But in reality, there's a very big gap between what I was trained to do and today, my every day, day to day activity. I would say that probably the most important training that I got during my days at the Weizmann Institute has got less to do with differential equations or molecular biology, and it was more about a story that my mentor, Professor Uri Alon, told me when I was three years into the PhD, about three years into the PhD, he asked me, Am I already in the cloud? He said what? And he said, are you in the cloud? I said, Well, what is the cloud? He said, Well, every PhD, every scientist, when you start your PhD, you know, you have you go you go and read the latest papers in Science and in Nature and you see how somebody starts at Point A, makes a hypothesis about point B and then take the straight line from A to B, and then you say, OK, I'm going to do the same thing and you start at Point A, the known. You shoot for the unknown and you start going and suddenly you hit a roadblock. And then you hit another one and another one. At a certain point, you'd really lose direction, which he called the cloud. You're in the cloud. And then if you have enough perseverance and luck, you find a point C which is not exactly where you thought you're going to end. You go there with, you know, your last energy. And if you're lucky enough, then you publish another paper about how you started at point A, went to point C and connected between the two dots with a straight line. And then you have another generation of PhDs that are asking themselves, Well, why am I the only one that's struggling? And that lesson about how to be in the cloud, how to deal with uncertainty, to deal with failure and still move on. That is probably more important in the training that I got to become an entrepreneur and CEO of a company than any specific scientific knowledge.

Harry Glorikian: Ok. Yeah, no, I mean, trial and error, dusting yourself off, getting up and moving forward is, you know, my wife calls me crazy when I keep doing it, but I think you have to be a little on the edge to constantly keep repeating and being willing to fail and then stand up and then move on. Maybe it's a, I think I was reading a paper recently that said forgetting quickly is evolutionary, you know, a positive trait so that you forget what happened, that wasn't good and you keep moving forward. So. But let's talk about your company, MeMed, like you started that in, I believe, it was 2009. And what was your founding vision? I mean, if you can talk about what you and your co-founder did when you came up with this idea, I think you were both studying at the Technion at the time?

Eran Eden: Yeah, so so he was studying at the Technion, M.D., Ph.D. I was studying at the Weizmann Institute and Data Science and Biology. And frankly, I would love to tell you a story about a vision, but it started with a game. I don't think we had the presumptions to have really something that would grow to what MeMed actually became today. It was playing. We both have had different reasons first of all for doing this. I can say that from my my end, it was probably a pretty big gap between the places, the caliber of where we were able to publish high impact journals. And when I was looking at myself in the mirror and I was asking myself, Is this actually going to have an impact on real patients? I couldn't really see the connection. There's another reason why I decided to found MeMed or co-found MeMed. That's probably off topic for today. We can take this on a beer some time when we meet face to face. But so it's first of all, it didn't start with a vision. It started with a scratch wanting to apply a some of the know how that we had had in converting between molecular immunology and data science, and to try to solve big, ugly problems that don't have a good solution in 21st century medicine and trying to find something pragmatic now rather than having it a eureka moment. You know, some pioneers describe a eureka moment where suddenly you have the best and coolest idea in vision. For us, it was darkness for almost a year rather than the eureka moment. It is was more like an evolutionary process. Trial and error. We tried a bunch of solutions to problems that didn't really exist until eventually we came up with what we want to work with, but again was no, no eureka, and the way that it actually started was again, Kfir was coming from from med school talking about this problem of of AMR, antimicrobial resistance and the problem of distinguishing between bacterial infections and given our different backgrounds, we said that's interesting. How can we apply immunology and then science to try to solve that, and then at that point, we formulated what was to become MeMed's vision. And MeMed is based on a very simple premise, a very simple idea. Our immune systems have evolved to tell us what's going on our bodies and all we do at MeMed is we listen to the immune response with biochemical sensors and machine learning and what have you. And we use that to translate or decode the immune system into insights that can potentially transform the way that we manage patients with acute infections and inflammatory disorders. The first problem we went after, because that's a very lofty goal, was potentially the most prevalent clinical indication on the face of this planet. A child with sniffles. Our elderly patients that coughs. Come to the doc, they have a fever. As a parent, you're many times hysterical, you're asking yourself, is it a bacterial infection or bowel infection. If it's a bacterium, antibiotics. It's a viral infection, chicken soup. And we said, Well, what if we can harness the immune system? What if we could measure or listen to the immune system in real time and use that to try to aid clinicians to tackle this seemingly simple problem? So the vision was listening to the immune response. In the first embodiment of the first problem we went after is this huge intractable problem, B versus V versus. Bacterial versus viral infection. To treat or not to treat.

Harry Glorikian: Yeah, I mean, you know, it's funny, you say simple, and I've worked in this area for a long time and now not simple, not simple, but I've been watching dozens of companies over time try and tackle this problem, and everybody always comes at it and says, Yep, we should be able to do it. And I'm like, OK, that's a big hill, you know, to go and try and die on so. But you got FDA approval for your device in the U.S., and I want to talk about that later. But it did take 13 years. Like to, you know which parts of the process turned out to be harder or slower than you thought it would be?

Eran Eden: Before I answer that, I just want a minor correction. I didn't say it's simple. I said it's a seemingly simple problem. In reality, it's an extremely difficult problem to go after. I think some of the most the biggest challenges that we have can be phrased in a very simple manner. But as you alluded to, yeah, it's an intractable problem. Bacterial and viral infections are often clinically indistinguishable. And it took us over a decade to take this from my idea on a napkin and grandmother's kitchen. That's where we found with no garage, it was Grandmother's Kitchen to what is considered a landmark FDA clearance that I think many folks did not believe we're going to be able to get this because it required so many innovations, not only on the technological side, but also on the regulatory side. And when you ask why only a decade? I think it's, we're very lucky that it took us only a decade and it sounds there, let's not call them challenge. Let's call it problems. Challenges is something I always envy the people that have challenges. We have problems with immune, and we work every day to solve those problems, right? So. So there's many problems or hurdles you have to go through. So there's first of all, you have to overcome some pretty big research issues, where do you find these hypothetical molecules of the immune response that go after bacteria and viruses. So research, then you learn the hard way.

Eran Eden: The research is very different from development, and development is very different than product, and product is very, very, very different than manufacturing, and manufacturing is very different than regular regulation, and regulation is very different than reimbursement in marketing, which is a very different than commercial, et cetera, et cetera. So it's not good, it's not enough to excel in one thing. You have to really reinvent the wheel on several things, and as a company and as a team, reinvent yourself, and that's probably one of the biggest challenge, probably your biggest impediment to progress is yourself and your team because you might be an excellent data scientist, but you have to talk with the clinician. You might be an excellent clinician, but you have to talk the language of the molecular immunology. You might be very versed in all these three, but it's still not product and it's still not the graphical user interface. And how is that connected to manufacturing and really creating a culture or a team that can combine these seemingly very diverse elements within a small company. That is a very, very daunting and big task, and again, we frankly failed on multiple avenues there. We had to go back, we were in the cloud and we had to reinvent point C again and again and again. So, you know, we were in a very far position that we are today that we thought we were going to be at this stage.

Harry Glorikian: So I'm going to ask at some point, you know, after this whole interview is I'm going to encourage you to write the next IVD book because everything you said is absolutely the way that I've seen it over time is, you know, having to bring all these pieces together is not trivial in our world. But let's step back here for a second for everybody that's listening, right? Talk a little bit about basic immune system biology and the, you know, technology behind your diagnostic system. So if someone presents with an inflammatory response, why is it so hard for doctors to destroying distinguish between the bacterial and viral infection?

Eran Eden: Because bacterial and viral infections are clinically indistinguishable and you don't have to be an M.D. to to understand this. Intuitively, we know our kids so well. But still, you know, when they have a fever or runny nose, you know, we know that it's 80 percent, 85 percent a viral infection. But what if? What if there's a lingering bacterial infection? And it just it turns out that because of the clinical manifestation is very similar. It's really hard to figure it out. Not only children, also adults with suspected LRTI or a fever without sores, and even when we apply modern, I would say diagnostics, there's still a big gap that remains. So for example, when as a scientific community or a clinical community, when we approach this problem, we have tools at our disposal. A rapid strep test. A rapid influenza tests. Multiplex PCR. In today's world, I think everybody, even my grandmother is talking to me about the difference between rapid antigen tests suddenly becomes a really interesting topic over, you know, weekend dinners, culture. So there's technologies out there. And going back to your question, why is it still, why is there still a gap? And we've identified several things. The first one is probably the most trouble is time to results. Many of the clinical encounters, you want to have the solution here and now where whereas that technology that we have often provide solutions in hours and even days, and that's not always good.

Eran Eden: That's one hurdle. Not the biggest one. The second one is that many times the infection site is inaccessible. Take, for example, otitis media, an ear infection or sinusitis or bronchitis or pneumonia. It's really hard to reach the infection side and therefore identify the pathogen. It's one in four patients in the most prevalent disease on Earth. That's really hard. Third, even if you use the best, most broad technology diagnostics to try to identify the bug, say a multiplex PCR. In more than 50 percent, five, zero percent of the cases, you're not be able to put your hands on any microorganism, but you still, as a clinician, have to make a decision. And lastly, there's the issue of colonization. So even if you're lucky, the infection that is readily accessible and you do get some sort of a virus, for example, that you detect, say, for example, an influenza or or let's take a rhinovirus, the rhinovirus is very prevalent in children. That's a problem. It's very prevalent in children. Even if you take seemingly healthy children in a very high percentage of those children, they're going to have a rhinovirus. So mere detection does not apply causality. All this complexity is sunk into this few minutes that the clinician basically needs to make a decision, and it's a really hard dilemma because it's hard to know to distinguish between the two and the ramifications could be quite significant.

Harry Glorikian: So I know the answer to the question, but I'm going to ask it so you can explain it is: So who cares? I mean, I know that it's ineffective to treat a viral infection with antibiotics and that only you know, that only work against a bacteria, but you know. We've been doing a trial and error, so what's the downside of doing that?

Eran Eden: So it's actually a pretty deep, it's a very deep question because there are several layers. You're right, this sometimes people actually say there's several layers to answering because the first one is, well, if you treat erroneously, with antibiotics, antibiotics, because of this uncertainty, there's a lot of antibiotics overuse that one of the consequences of this it drives anti microbial resistance, which basically means that the drugs don't work anymore. And if we continue on that path, we will potentially lose modern medicine because if you lose the potency of antibiotics, you cannot treat infants when they have an infection. Or an oncology patient that would succumb to a parasitic infection, or even yet have your wisdom tooth pulled out, because antibiotics won't be effective. And there's several quite influential studies that came out in the last few years. The last one actually in The Lancet came out two weeks ago portraying a world without antibiotics, which is, you know, we're seeing right now the consequences of COVID SARS-CoV2. Some might argue it's not a smaller problem. So that's and it has both clinical and health economic consequences. According to Jim O'Neill, over $100 billion by 2050 in lost GDP.

Eran Eden: And. And it's a big number, right? It's a really, really big number. And maybe, maybe it's overly inflated and maybe it's conservative, but it's a big problem. The issue is that nobody cares. Sometimes the individual doesn't care because the doctor, right now, when he has a patient in front of him, he doesn't think about the masses. He thinks about the patient. So you might ask, well, what the doctor care. Why does the patient care? And it turns out that there is an angle on the personal level as well, not only the societal level. If you give erroneous antibiotics, you can drive anaphylactic responses. You can drive allergies, which have a toll. But then there's another element that people are less aware of. In addition to overuse, there's also simultaneously underuse. One in five patients that have a bacterial infection are not receiving antibiotics in time. There's much less publication on that. But it is a reality. And that also has consequences, including prolonged disease, duration and sometimes even morbidity and mortality. So it's a really delicate equation, right? You don't treat. And you don't want to get ... some countries overtreat, some undertreat. And again, at the end of the day, the day to day, it does have ramifications both from the patient and on the doctor.

Harry Glorikian: You know, if we could accurately treat people right, I think there would be a whole host of issues that could get solved and a whole host of issues that wouldn't emerge because of overtreatment or treating the wrong people that you know, we could spend hours over a beer discussing the microbiome and allergies and all sorts of other consequences of doing this. 

[musical interlude]

Harry Glorikian: Let’s pause the conversation for a minute to talk about one small but important thing you can do, to help keep the podcast going. And that’s leave a rating and a review for the show on Apple Podcasts.

All you have to do is open the Apple Podcasts app on your smartphone, search for The Harry Glorikian Show, and scroll down to the Ratings & Reviews section. Tap the stars to rate the show, and then tap the link that says Write a Review to leave your comments. 

It’ll only take a minute, but you’ll be doing a lot to help other listeners discover the show.

And one more thing. If you like the interviews we do here on the show I know you’ll like my new book, The Future You: How Artificial Intelligence Can Help You Get Healthier, Stress Less, and Live Longer.

It’s a friendly and accessible tour of all the ways today’s information technologies are helping us diagnose diseases faster, treat them more precisely, and create personalized diet and exercise programs to prevent them in the first place.

The book is now available in print and ebook formats. Just go to Amazon or Barnes & Noble and search for The Future You by Harry Glorikian.

And now, back to the show.

[musical interlude]

Harry Glorikian: So your system, which is, I love, is a basic blood test, right? So the MeMed BV looks at three immune system proteins in the blood: TRAIL, IP 10 and CRP. So how are these proteins related to infection and how can measuring their levels tell you about the nature of the infection?

Eran Eden: Ok, so. Each one of those proteins that you just mentioned plays a critical role in the immune response to different invaders, bacteria and viruses. What's special about this particular trio, is that they work really well as a team. Maybe if you take a step backward to identify them, we had to run for about four years what is arguably the largest prospective proteomic study ever to be conducted of the human response to acute infections. And start with a host of multiple tens of thousand proteins bioinfomatically and then down-select this eventually to three. And these three, while none of them is perfect in itself, they cover one another's blind spots. So let's go one level deeper. When we went on this, one of the things where we were surprised to find out that is a clinical community, we've been obsessed with the bacterial side of the equation. Every biomarker that you have in 21st century medicine, 20th and 21st century medicine, has been mostly predominantly upregulated in bacterial infection. Procalcitonin: bacterial infections, CRP: bacterial infections, white blood count: bacterial infections, absolute neutrophil count, which we use as part of routine day to day care: bacterial infections. What about the viral side of the equation? We couldn't find one that was used or cleared by FDA as part of 21st century medicine. The last. The reason the FDA cleared us, we actually just cleared the first viral protein ever to be cleared by FDA. And so we went on this fishing expedition and four years into the process, again, this was 2009-2013. We identified this trio. TRAIL Is a protein that shoots up in your bloodstream when you have a viral infection, whether it's a common influenza A, influenza B, parainfluenza or corona, and it has this very unique property that it goes down when you have a bacterial infection, why nobody really understands the reason. But it really creates a very dramatic full change because of this up and down type of a response. And the story there, there's a lot of interesting stories around TRAIL, but one of the ways mechanisms by which it does that it causes the cells that are infected by viruses to commit apoptosis. Cells suicide. And by that, protect the brethren cells. So that's one of the mechanisms that the body is using.

Eran Eden: The second one is called IP 10, which is an interferon. This protein basically shoots up in your bloodstream if you have a general infection, and more so if you have a viral infection. It recently got a lot of headlines in the context of SARS-CoV-2 and hyperactivity of the immune response. It's also associated with lung injury, but a really interesting biomarker that plays a critical role there in clearance of infections. The third one is called C-reactive protein, that's been around for about 40 years. Goes up in bacterial. And the nice thing about them? They work as a team. So as I said, they're not perfect. So take, for example, CRP. It's reasonably OK after 48 hours. But because it takes it to about 48 hours to reach maximum level, but in the early phases, you have a blind spot. Whereas TRAIL, at time of symptom onset, it's already differentiated, so they cover one another. By the way, we didn't identify this. The computer identified that. This is a lot of insights that we had in hindsight when we were looking.

Harry Glorikian: Yeah, that was going to be my next question, which is. You know, the the heart of the show is always like, you know, artificial intelligence and its whole basket of tools and biology. So how does machine learning come into this process? Is there some corpus of training data that shows that certain levels of these three proteins correlate? Or can you tell us, you know, how you developed that part of the system?

Eran Eden: Absolutely. And I think again, I was teaching a machine learning at the Weizmann Institute over a decade ago before it was a sexy topic. You know, people are using the term machine learning and data science so often so frequent. I think what's important to say is that machine learning is part of the component technology, but there's hardcore immunology and molecular biology. So it's not just one technology that we're, you know, it's a it's a very high entry barrier because of that and adds to the complexity. So that's one thing, just to put machine learning in context. Where machine learning plays an important role here is two places: in the development and in the final product. In the development, there's a process of how do you find the optimal team of biomarkers that would give you the the best performance? And over there, there's a lot of activities around using publicly available data sets and and proprietary data sets and data analysis and statistical analysis and feature selection and find the right models, et cetera, et cetera, coming up with what is the right model. Some of these are more conventional tests. Some of these are more cutting edge tests in the final product. It basically uses what's called a supervised learning approach, which basically means the following. Imagine every problem in here, I'm going to be a little bit technical here. Imagine you have, let's say one feature. Say a viral biomarker. TRAIL. High levels, viruses, low levels, bacteria. You find some sort of cutoff that separates between the two. It was the most informative biomarker that we found.

Eran Eden: Is it good? It's reasonably good, but there's no perfect biomarkers out there. We don't have them, nor do we believe they exist. Nor do we believe in unicorns, even though my daughter is trying to continually persuade me that there is one. Then you add another biomarker to that. Imagine that you have right now a two dimensional grid. And now suddenly, every patient is met this two dimensional coordinates and you have a cloud of bacterial and the cloud of viruses. And you find a line that separates the two. And then a third dimension and a fourth and so forth and so on. And eventually, the problem becomes how can I find this type of plane or hyper surface that separates between the cloud of bacteria and the cloud of viruses? This is the essence of the machine learning and the way you go about this. You train give it a lot of patients, a lot of data, and then you train the system. And the more data you have, the smarter it becomes. The same principle applies for doing diagnostics in oncology, span detection, computer vision and what have you. It's the same underlying, often similar underlying principles to try to solve many of these problems. So hopefully I was able to to simplify and somewhat exaggerate how this is actually working and where the AI plays here.

Harry Glorikian: So what's that accuracy rate of the diagnosis from your system? And is there are certain things, let's say it's good at in certain things, it's maybe not so good at?

Harry Glorikian: Yeah, absolutely. So so if you look at the overarching population, if you look at our pivotal FDA study, the AUC, the area under the operating curve, the entire population was 0.9 to 0.97 across different cohorts, which is considered very high. So that's the short answer. The more we see deeper level, it's there's obviously nuances across different populations. One of the things you have to show is what happens in children versus adults. Upper respiratory tract infections, lower respiratory tract infections, et cetera, et cetera. So we've shown a relatively consistent and robust response. That's how the system was developed. But there are, for example, certain viruses that we know that we perform less accurate. For example, adenoviruses. Adenoviruses are notoriously hard to to treat well. By the way, they're one of the most prevalent, for example, viruses in children, why? Because the immune response looks like a bacterial infection. For many, many reasons. So white blood count is going to be elevated. Procalcitonin is going to be elevated, CRP is going to be elevated and we're often going to overtreat with antibiotics. So if you look at our performance in that particular sub-cohort, our performance drops somewhat from, you know, a 0.90-something to maybe 0.89, but that's actually one of the viruses that we see the most added value because compared to standard of care, it's many times close to flipping a coin.

Eran Eden: So even though our performance is eroded in this particular virus. The standard of care in this particular situation is particularly challenged, and it's almost 0.5, 06. so that's one example. There's multiple examples. We can study the immune response to pathogens again for almost a decade now. This is just one interesting anecdote. And I think just connecting this to the who cares question that you had. So here's an interesting case that we had a few weeks ago. A child, young, three years old, coming to a major medical center, not really sure if it's a bacterial or viral. Ran a PCR, came positive for adenoviruses and it looked a little bit bacterial. But yeah, OK. Adenovirus explains everything. Released home. Got a 99 score. 99 probability of bacterial infection. So they start to do additional follow up and then it eventually turned out to be a bacterial axis in the spinal cord of that particular child. It had to be mechanically removed. This is a very dramatic case. This is one of those potentially underused cases that could be very dramatic. This is very rare. It doesn't happen often. But again, it's hard. It's really, really tricky to distinguish between infections and we added this right now, this is how everything maps together to the adenoviruses and and to why we think this could be helpful.

Harry Glorikian: So, you know, like I said earlier in the show, you know, you got FDA approval and granted 510K clearance back in September, which congratulations, that's a huge step. But you know, for everybody listening, what Gates does, does that open for you. What's the pathway to getting the device out into the market?

Eran Eden: So as you said, first of all, you have to get the clearance, which I think took us almost five years working with FDA. FDA, by the way, we've worked with them extremely collaboratively and they've been instrumental in helping us form and shape, what's the methodology to actually prove something. We didn't talk about this? But how do you prove that absence or presence of bacterial viral infection in the absence of a true gold standard? So let's put that thing aside. We were able to work with FDA and come up with a methodology to do that. So now, what is required to take it to the next step? There's several things. The first one you need, and we didn't talk about this, you need a way to measure those biomarkers. You need a platform. Right, one of the challenges that we had is that in the early days, none of the big strategic players, the Roches, the Abbotts, the DiaSorins of the world were willing to bet on this because the risk were so high, as you alluded to in the beginning, the graveyard. And nobody got FDA clearance, so they basically they wouldn't. They were not willing to bet on us today. Some of them become partners and we're working with them. So it's, you know, there's been great development. But at that time, it was really hard. The platform is also challenged because some of the proteins are picograms, some are nanogram and some are micro per mil, which poses again the challenge from a technological perspective. Again, not going too much into the technology side, but we've been able to come up with a technology or a platform that can actually measure multiple proteins across almost a six to nine order of magnitude range. And so you have to have a platform and can do that in about 15 minutes right now, serum working in whole blood.

Eran Eden: The second thing you need, you need obviously manufacturing capability, which is again, a different story, you have to manufacture the cartridge. The third thing you need is building the clinical evidence beyond, I mean, FDA's great, but you have to create what's called a clinical utility, real world evidence, what have you, working with peers. Work with partners or with clinicians working the societies. Publishing. Building a commercial team which we're right now established commercial team in the U.S. So there's multiple things. And probably last but not least, this is too big of a challenge to go at it by yourself. You need to have partners. Whether it's governments, the U.S. Department of Defense, the European Commission have funded this heavily and have been amazing partners, whether it's strategic partners, you can take it by yourself versus vs not one market. It's markets. You have patients in the wards and the EDs and the urgent care physicians' offices, retail clinics. No single player has enough of a presence in one platform that covers it all. So again, we've announced about a year ago, you know, the first partnership with DiasSorin, which has today almost a thousand installed installed across Europe and the US in these large automated immunoassay machines. And that's covering certain parts of the market that are overlapping or, sorry, that are complementary where we're going at. So that's a little bit of what needs to be done. But again, it's changing the boundaries of what what we've been doing so far, and that's always a it's always a challenge, but also an opportunity.

Harry Glorikian: So you guys raised I believe it was $93 million, if I remember the number correctly, in new funding, which sort of really adds to the firepower of being able to take that next step, but. You know, can you can you envision a future where we get a solid diagnosis and an appropriate treatment plan, you know, quickly while you're in the doctor's office?

Eran Eden: Oh, yeah.

Harry Glorikian: That was the shortest answer you've given yet.

Eran Eden: I think you can be much more radical. I think there's several things that are happening. There's two major discontinuities. There's a technological discontinuity. There's a regulatory discontinuity. And I'll actually add another one, which is there's a psychological discontinuity. The technology that we can do today that we have today, the tip of our fingers can do can provide so much valuable information that can help make better decisions. The regulatory framework has changed because of COVID, it's basically shattered a lot of things. And from a psychological perspective, I think there's a push to polarization, right? Both decentralization and centralization at the same time. And so I definitely see that happening. I think we can take this several step further. How can we take it from physician's office, also retail clinics and even further? And that will take time, obviously, because we're dealing here with some pretty, pretty deep questions. But I think the world across multiple fields and this is not different than anything else. I think it's definitely going in that direction.

Harry Glorikian: Yeah, no, I mean, I was going to, you know, looking at what you've created and, you know, obviously first getting everybody on board, but then seeing how it can be deployed at a CVS or something like that, it could drum, you know, you could have a dramatic impact on how we manage patients. The whole antibiotic dynamic and maybe even relieving stress on the system so that, you know, it doesn't get overwhelmed by what your system may be able to sort of help get to a much faster, much more accurate answer too.

Eran Eden: I wanted to say relieving stress from stressed mothers and fathers. But yeah, I agree with you also, relief. And by the way, as you start going from more centralized to decentralized, there's obviously additional workflow challenges. How do you make this simpler? There's regulatory bars that you have to meet. How do you go from a mod complex to a clear waived test that can actually go to those directions so that there's we still have we have work, there's work, work to be done. But I think we've been able to potentially break a glass ceiling in terms of getting the clearance. And now I think with that, there's going to be additional innovation that will come in both by us and others who are entering the space.

Harry Glorikian: So. Just slightly moving to one side is like, how has MeMed responded to say, COVID-19? I think you guys have developed a test that runs on your platform and predicts how severe the infection will be. How does that test work? Did your previous work, you know, and also did your previous work like on the platform prep you for this virus? Just curious how it works and how you got there?

Eran Eden: Absolutely so. So it always starts with the clinical question. I mean, many of us are technophiles, but at the end of the day it's about solving a real problem. And the problem here is the following. You have see SARS-CoV-2 positive patient presenting to the ED. And one of the questions that we have in mind is whether that patient is going to deteriorate or not. Do we escalate treatment or not? And it's a real question, right? And the more you know, the more stress the system is feeling because, you know, because of the the peak of a pandemic, the more important that is. So we said, Well, how can we harness the technology? Is the framework that we created host response in general, right? The biomarkers we've developed, the platform that we have, the Biobank and what have you. And so and how can we take a process that maybe took 10 years and now collapsed into something maybe that's 10 months? How do you get a 10 X? And and first of all, with amazing partners around the globe, you start running huge clinical studies to basically collect patient samples. We also use again information from the public domains, our own repositories, our own previous data because from many perspectives. Sars-cov-2 is very interesting, but guess what? Similar to SARS and to other types of severe viruses, there's differences, but also commonalities.

Eran Eden: So we use a lot of the bioinformatics, the previous data samples. Current data samples. The know how and the platform that's readily available right now. They can measure basically anything to collapse this and develop. This is probably just got clearance in Europe that basically allows to take a snapshot, the main response again in real time. Give you a risk stratification regarding the probability of a patient to experience severe outcome defined as ICU level of care and mortality within two weeks. Again, it's only clear right now in Europe, not cleared in the U.S. and we view this also as a stepping stone going beyond just COVID severity to a general severity signature. So what you do, both B versus V and severity, what if you could do it in the same cartridge or what have you? So I think what's what's really interesting, we build here this core technology. We went after one big problem, B versus V, but now that you have that, you're like a child in a candy store. There's many more things that you can do. And rather than taking you a decade, you can start to collapsing things because there's a lot of there's a lot of. Resources that you can now leverage or platform that you can leverage, so that's a story around MeMed and COVID severity.

Harry Glorikian: Yeah, platforms are wonderful in that way, right, that I like a platform more than I like, like a, you know, sharpshooter bullet, from an investment perspective. Thinking about it that way. But so. You recently got COVID. We were supposed to talk like over a week ago, and I, you know, we had to postpone it. Did you use the test on yourself? I mean, if you did, like did it work the way you thought it was going to?

Eran Eden: Yeah, so so yeah, I got my I got it from my daughter. We went on a trip and five out of five family members got infected. So yes, it was at least from our small experiment. It was very infectious. We got the Omicron. Actually we didn't have symptoms, apart from the fact that I think it just jacked up the energy level of my kids. So before we talk about running around the house and thank God, you know, my wife didn't didn't kill any one of them. So there's no casualties from this, from this infection. So because we didn't have symptoms, we didn't go to the ED. It was not relevant. You have to have SARS-CoV-2 symptoms. So in that case, no, no, no need. I mean, we were pretty much hunky dory. But what I can tell you is that on the B versus V. Again, it's potentially bacterial and viral infections are potentially the most prevalent indication in children. And my children, those little incubators of bacteria and viruses, are no exception. So I had a chance to use the technology on my kids many, many times, including last time was about a month and a half ago, and my eldest daughter, who is four, before going to a business trip. And my wife is asking, is it a bacterial infection? I said, I don't know. She spits on me. The shoemaker is going barefoot. So we ran it. It was a viral infection. No antibiotics. Went back to school. So and I got a lot of brownie points with my wife and my mother in law, which is obviously always very, very strategic.

Harry Glorikian: That's that's a good one. That's always helpful. Exactly.

Eran Eden: So we're actually using this quite often in our families as well. And it's very very gratifying.

Harry Glorikian: Interesting. Excellent. So now you guys are, you know, I believe you have an office in Haifa, which I remember as being beautiful and hilly and wonderful food, and then you have Boston. You know. What are the strengths of being in these two locations. What happens in Boston that can't happen in Haifa and vice versa?

Eran Eden: Well, again, we're going after a global problem and you have to have a global team to have a global perspective. Whatever you have in San Francisco today, you have tomorrow in Shanghai and the day after that in Tel Aviv. So you have to look at this from a global perspective, number one. Number two, since the US is the primary market, as I said, we have to build a very significant presence in the U.S.. Why specifically Boston? Very talented pool of, a pool of talent that's very wide in the domain. There's a big overlap in terms of hours between Boston and Tel Aviv, so you can grow one unified team. And that's basically, that's where we're basically building our U.S. headquarter. And the team is quite complementary. Again, we've we've recruited by now roughly 25 to 30 folks, folks with a very strong background, both IBD, Troy Battelle, formerly Thermo Fisher, who's buying commercial for microbiology in the Americas. Fred, who is running Corp Dev, from bioMérieux. Again, another large multinational, Jim Kathrein was former head of sales for BioFire. Again, not sure if your audience is familiar with but and so forth and so on.

Eran Eden: And Will Harris was running our marketing global marketing, is ex-Amazon and then before that, 15 years in IBD. So it's really bring here a blend of, we call this affectionately an anti disciplinary team. We don't care about disciplines, we care about solving big, ugly problems. So you have to bring the IBD experts with the clinicians, with the folks and the big data science side or in the molecular immunology and the manufacturing. And nobody knows, single location has all the know how, no single location has the recipe because frankly, we're doing here something new. There is no real tech like this. And so bringing those this pool of talent, I think has really helped us, propels us moving forward. And it is the bridge to be able to to launch in the U.S., a U.S. very focused, commercially marketing product where a lot of the I would say more of the molecular immunology data science team is more in Israel. I'm simplifying and exaggerating. That's some of the team.

Harry Glorikian: So the last funding round, was that the argument to the investors, like we need to hire these types of people to help blow this out? What was what was the rationale for that last round?

Eran Eden: So, so basically three things. Number one, commercialization. U.S. Europe, Israel. That's our initial focus and then the rest of the world. Second is product pipeline, so we talked about bacterial versus viral infection and a bit about COVID severity. But what would you do if you had a blank canvas and these platforms to go after the new response to measure the immune response? What additional big problems would you go after? So it turns out that there's some pretty interesting stuff in. We're working on additional activities. So that's the second thing product pipeline. And the third thing is a scaling manufacturing. So as I think people have a new appreciation for manufacturing and supply chain during COVID times, it's a really big topic and critical for success. So this these are the three major elements that we raise the funds for.

Harry Glorikian: No sounds I've I've been I've seen this rodeo a few times, so yes, I understand completely. So, well, you know, especially because I come from the diagnostic world and I can't wish you enough success because we need more products like this out on the market to help us manage patients and help give physicians better information so that they can make better decisions, right? More informed decisions than just, you know, looking at a patient and trying to figure out what's going on. So I wish you incredible success and I'm, you know, great. Great to have you on the show.

Eran Eden: Thank you so much for for the kind invitation. Enjoyed our discussion.

Harry Glorikian: Thanks.

Harry Glorikian: That’s it for this week’s episode. 

You can find a full transcript of this episode as well as the full archive of episodes of The Harry Glorikian Show and MoneyBall Medicine at our website. Just go to glorikian.com and click on the tab Podcasts.

I’d like to thank our listeners for boosting The Harry Glorikian Show into the top three percent of global podcasts.

If you want to be sure to get every new episode of the show automatically, be sure to open Apple Podcasts or your favorite podcast player and hit follow or subscribe. 

Don’t forget to leave us a rating and review on Apple Podcasts. 

And we always love to hear from listeners on Twitter, where you can find me at hglorikian.

Thanks for listening, stay healthy, and be sure to tune in two weeks from now for our next interview.