Dr Rupy: Eran, we're going to get right into this. So, everyone's heard of diet pyramids, eatwell plates, general uniform advice. What do we get wrong when we create those templates and we promote it to everyone in the population?
Professor Eran Segal: I think the main thing that we get wrong is that we're trying to fit the same diet to everybody. And our study and other studies basically show that people have very different responses to food. And when you see that data, then it just shows you, just the data shows you that you really have to tailor diets to people in a different way because because we each are different, we have different composition, different genetic composition, different bacterial composition. We know that our gut bacteria play a major role in our health. And so we're just we're just different people and that's why also the food that is best for us should be different. I think that's the fundamental thing that we get wrong and I think it's never going to be the case that we'll be able to find a single diet that fits everybody.
Dr Rupy: So even the general sort of parameters around healthy eating, do you have an issue with that or is it the minutiae of things that we can tweak, maybe the 10% or something like?
Professor Eran Segal: Yeah, so so I think it's more than 10%, but obviously I would agree that there are things that are probably universally not good for us. So we know that trans fat, for example, is very bad for us. We know that we we did studies actually on non-caloric sweeteners. And we showed that actually those alter gut bacteria in a way that when you take that gut bacteria and you transfer them to people, they can actually generate symptoms of diabetes in mice. Yeah. So this we published in 2014, it made a lot of headlines. Since then, that study was actually reproduced several different times and and now more and more people are realising that for example, those ingredients are also probably bad for us and they're gradually being recommended against.
Dr Rupy: When you see those studies from 2014, the ones that you were involved in, and manufacturers are still using similar sweeteners, maybe even sometimes the same sweeteners. Does that annoy you? Does it does it sort of irk you the fact that it's been nearly 10 years and we haven't really moved the needle in the direction of actually helping people?
Professor Eran Segal: Of course it annoys me, but I also understand that companies that are selling in billions of dollars, and we can talk about different areas that we also did research in, for example, probiotics. I think probiotics as a as a direction, as a concept could be a good thing, but not the probiotics that companies are selling today, which is a multi-billion dollar industry, but just the probiotics that were chosen were not chosen because of studies done to see the health benefits of these probiotics, but just because by happenstance we found some bacteria that are easy to grow, they don't affect the food too much, they don't change the texture, the smell and and so on. And but they're sold in billions of dollars. So I think in the end the regulators would have to be involved before we will see a major change in that. And we are seeing the regulators gradually coming in. It's a very slow process. I think the analogy is if you look at what happened what happened with cigarettes, the regulators they didn't ban it, but they increased taxes so much that and we do see a major decline in cigarette intake. So I think we'll have to see activities of regulators before we see we see a change. Just recently the World Health Organisation actually declared that some of the non-nutritive sweeteners, in part based on our studies, in part based on other follow-up studies that I mentioned, are actually recommended against. So I think these these these things will trickle in and eventually have an effect.
Dr Rupy: And I guess in the meantime, you know, educating people as to what they can do to minimise any particular risk. And I think what regulators and perhaps manufacturers sort of turn a blind eye to is the precautionary principles around food and introducing it before, you know, they've actually tested it and actually seen its safety profile. Let's dive into one aspect that I think a lot of people know your work for, which is everyone's individual response to sugar. Why don't we talk a bit about how there's immense variability in someone's sugar spikes after eating the same products and some of those unlikely products that you think everyone would have a sugar response to, but you actually see sometimes it's not as bad as as we would be led to believe.
Professor Eran Segal: Yeah. So, of course, but maybe before that, a few words on even why blood sugar response to food is important because I think people need to understand that. So when we eat food that contains carbohydrates, our body digests that food and releases it as sugar into the bloodstream. From there, our body signals with insulin to our cells that there is now sugar that they can uptake and use as energy. But if we eat in excess, that excess sugar is eventually converted into fat in our cells and this is actually the primary mechanism by which we gain weight. It's not by eating fat, but it's actually by eating more carbohydrates that we're not using as immediate energy but storing them eventually as fat. So it's really key for weight management. It's also key for development of diseases like diabetes, even cancer and and some other diseases because as we burden our system with more and more spikes in blood sugar levels, our body then needs to loses some of the sensitivity to insulin. It needs to then secrete more insulin to signal the cells to uptake the same amount of sugar and eventually our beta cells are not able to produce the amount of insulin that's needed and that's when your beta cells die and you develop full-blown type two diabetes. So so it's really something key which is why we decided to study it instead of just looking at for example, changing weight after dietary recommendations. Because because it's important and because if you look at changing weight, that's a single measure, it takes weeks to affect and you can't then link it back to every individual meal. The nice thing about blood sugar levels is that you can track it continuously with these devices now with these sensors of continuous glucose monitors and you can get a quantitative measure, a health measure of your response to every individual meal.
Dr Rupy: Yeah, absolutely. And I think even on a micro level, so the macro picture of these glucose excursions, particularly if they happen over time and your HBA1C is reflective of that, that can lead to weight gain and the other issues around insulin resistance. But even the short-term glucose excursions, do you lean into the the idea that those excursions outside of what the normal parameters are can lead to inflammation, advanced glycemic end products?
Professor Eran Segal: Absolutely, absolutely. And I think that's been shown more and more and they can they also affect your blood vessels and your your vessel health, which is also super important in the in the long run. So yeah, I think we need to as much as possible avoid these spikes. Now, this is not to say that live your life, of course, and sometimes enjoy things that even spike your glucose levels. That's fine, but you know, the the less you do that, I think the healthier you will be. And so this is this is one aspect that we really want to to affect.
Dr Rupy: Yeah, I'm glad you said that because I'm not puritanical about that at all and I'll definitely eat a doughnut or like, you know, we were just talking about some nice restaurants that we're going to experience in London. So with that in mind, we understand sort of the basic picture around sugar control is important. How did you demonstrate the the immense variability in the glucose responses?
Professor Eran Segal: Yeah, so initially when we started this, um, actually I got into this from personal interest. Just I wanted to find a diet that's good for me and reading all the being a scientist and reading all the diets and seeing the recommendations change every month, it was clear to me that there's not really solid science behind them. So I started experimenting with myself and I saw to my great surprise, now now it's less of a surprise, but back then, that when I ate just chocolate or ice cream, my blood sugar levels just didn't spike.
Dr Rupy: Oh, you were one of the lucky ones.
Professor Eran Segal: So so that that greatly, that was a shock. And after the first day, I remember telling friends that this is this is probably not true, so I'll do it again. And I did it again and again and again, and then I see the same response. And so and then somebody else in the lab started experimenting and he had opposite responses. So then we knew that okay, there is this is a surprise, we have to study this on a large scale. And so what we did was recruit a thousand people and connect them to these devices, the continuous glucose monitors, the CGMs, and then just have these people log on a diet app that we developed all their food intake. And just to standardise things, we also provided ourselves the breakfast, a very simple breakfast, just either four slices of bread, four slices of bread with butter, or just pure glucose. And what we saw was actually what we saw anecdotally on a few people, we saw on the scale of a thousand people. Basically, you give the same people different, you give different people the same food and it's true for every single food that we ever tested, you'll see very different responses. Some people will have almost no effect to that food. They'll eat four slices of bread, you'll see their glucose levels and if they didn't tell you they ate bread, you wouldn't know it from the CGM. Whereas others would have huge spikes in glucose levels, even more so than they would then those same individuals would have if they eat whatever pizza or chocolate or ice cream. Yeah, yeah. Um, so so this was this was shocking and again it showed that if we want to affect blood sugar levels and we just talked about why that's important to do, then we have to tailor these diets individually to people.
Dr Rupy: Yeah, so just to reiterate for the listener and the viewer, same same foods, drastically different responses. And we're talking about pure glucose in in some of the
Professor Eran Segal: Including pure glucose, but also, you know, complex carbohydrates. Sometimes foods that are considered to be good that would be recommended by a dietitian. So if you if you go to a dietitian and they recommend for you a Mediterranean diet and some of the meals that will include rice, maybe they'll tell you, okay, eat brown rice as your carbohydrate. We see that that brown rice spikes blood sugar levels for people, some people, much more than junk food. So if you look at that parameter, again, I'm not saying that then the junk food would be good, it may be bad for other things, but but at that but from the perspective of that parameter, and if you think you're eating healthy and you eat that day over day over day, you're basically that brown rice actually may be driving you to developing diabetes faster than if you didn't follow the recommendations of the dietitian.
Dr Rupy: So putting this into context because you know, people have come across glycemic indices for different foods. It's very well taught by dietitians, well-meaning nutritionists as well in general practices across the country, across the world. How do we generate those and clearly if what you're saying is true, those are deeply, deeply flawed.
Professor Eran Segal: So so first, how do we generate them? So typically, we do studies on a small number of individuals, 10 or 12 or 20 people, sometimes a bit more. We give them the same food, we measure their blood sugar response exactly as we did in our study. And then we take the average. So when you do that, of course, you will get a result and it will be true that some foods will have a higher glycemic index than others. What that means is that the average of those 20 people that you measured is going to be higher than the average for even those same 20 people when measured on a different food. But, and this is the key point, if you look at the average, of course, the average of 0 and 20 is 10, the average of 10 and 10 is 10, but in the first case, you have huge variability between the individuals, in the second, you have identical responses. And so what we're seeing is that in every single case, the glycemic index, when you look at the individual numbers, they'll be very variable. So what does that mean for you as an individual? If you don't know if you're one of those that contributed the low number or the high number to the average, of course, you don't know if that particular food is good or bad for you. So that's why, yes, you can measure the glycemic index. It's also true, this is the average of the food, but because of the very high variability, which is greater than the differences in the averages, it's very hard to put together a diet that would be good for every individual based on the glycemic index.
Dr Rupy: Do you think given everything that you've been studying for this amount of time, we should completely disregard these indices or is there any place for them in dietary guidelines?
Professor Eran Segal: Yeah. So, you know, I would say in the absence of any information, of course, if I was given the choice between two different foods, one with a high, one with with a low glycemic index, I should probably eat the low glycemic index food. Statistically, just your average person would have a lower response. So it's like what we talked about before. I I do believe that there are general principles that we can follow and that would be better than not following them and they would be good dietary advice, but I think they achieve only a very small fraction of the full benefit that we can get if we fully personalise diets for people. I'll give you an example. Of course, when we talk about all these blood sugar responses to food, the major factor that even we find in in our studies is that the amount of carbohydrates you have in a meal is the single most determining factor for the response. And and that's obvious, yes. Of course, even though we see variability across people, of course, on average, the more sugar, the more carbohydrates you eat, the higher the blood sugar response will be. But if you quantify that, then we see that the amount of carbohydrates in the meal explains about 15%, 15% of the variability in blood sugar response compared to the algorithm-based diet that we developed, which can explain 50, 50% of the response. So we get more than triple the effect of what we can explain in people's response when we just follow when when we take the full algorithm diet compared to only taking the amount of carbohydrates. And and this by the way is also very critical because if you look at a disease of diabetes, but type one diabetes, where the patients really don't have beta cells and they really need to inject insulin in order to lower their blood sugar responses, the standard of care right now is to tailor the amount of insulin to inject according to the amount of carbohydrates that you ate in the meal. And we know that that's off, that'll explain only 15% of the variability. And and what happens, and you see this in type one patients, is that sometimes they'll inject too much insulin, which actually could be very detrimental, eventually even leading to death. And sometimes they won't inject enough and their blood sugar levels won't go down and and that's bad for other reasons because high blood sugar levels eventually are very bad for your health.
Dr Rupy: Yeah, I I I mean, just leaning into my clinical practice, I remember vividly having some patients that really struggled to get on top of their blood sugars despite being pretty rigorous about their foods and ensuring that they were of the correct GI index and they understood the dosing of insulin. And so that alone can explain a lot of the variability that these poor patients were experiencing.
Professor Eran Segal: Absolutely.
Dr Rupy: Yeah. And hypos are nothing to be scoffed at. A lot of people don't realise that, you know, hypos can be extremely detrimental and the reason why a lot of them are admitted to A&E. Is there a putting that aside, which I think is a very important aspect of medicine and to understand the variability within people's responses. Is there a danger of being a little bit too myopic about glucose responses per se? I.e., you know, if I game my diet to ensure that I have a nice flat glucose curve, am I inadvertently going to be putting myself at risk by having too much saturated fat and increasing my risk of cardiovascular disease that can only be demonstrated over a longer period of time when I have a heart attack at 20 years because I've been eating fat the whole time.
Professor Eran Segal: Yeah, so so that's a good question. Um, so and I'll answer that in two different ways. I'll say first, in terms of pushing you more towards eating fat, we actually haven't seen that have a detrimental effect. In fact, in the studies that we conducted, we conducted randomised clinical trials on our diets, which do push people towards on average eating more fat. Of course, it varies between people. We've actually seen measures of fat go down, triglycerides, cholesterol, they actually go down. And the reason is because in a healthy person, even if you eat more cholesterol in the food, your body is able to clear that away and it doesn't elevate cholesterol levels in the blood. So I think for some of the aspects, at least we haven't seen any aspect, any any biomarker that we could measure that on our diets which really target blood sugar responses to be flat, we haven't seen any detrimental effect. However, having said that, I will say that of course, the goal can never be only to lower blood sugar levels because if you want to do that, then you should fast, right? If you fast, yes, that will lower your blood sugar levels, but after a month or so, you'll die. Okay? So, um, of course, that cannot be the only goal and we want our nutrition to be nutritious. We want to be intaking a variety of different food, diversity of foods, eventually all the minerals, all the vitamins that we need. So, so we have to have a diet that um, uh, tailors to all of those aspects. Again, we went after the blood glucose response, um, well, because it's clinically important, but also because it was the only thing, I think it's still the only thing that we can really measure continuously and really take an unbiased data-driven approach to that. I'd love for us to be able to measure inflammatory markers on a continuous basis, um, fat lipids, uh, and so on in the blood, uh, but we just don't have the technology to be able to measure that. On the day that we will, we'll be able to do exactly, take exactly the same approach that we did for glucose to looking at many other aspects. And until we do that, I think we just have to diversify our foods while maintaining low blood sugar levels.
Dr Rupy: Yeah. On the on the subject of biomarkers, I really want to get to the human phenotype project. Tell us about what that is and why you're doing it. Because compared to what we've just been talking about now, the stuff on glucose is like a drop in the ocean compared to the ambition of this particular project.
Professor Eran Segal: Yeah. So, so we started with this nutrition project for the reasons I mentioned. When we finished it, we said, well, we took, we looked back and we said, we were able as a lab to recruit a thousand people, get them to wear the continuous glucose monitors and really develop a product. So or develop an approach that we showed has beneficial effects. And so then we said, well, if we did that for nutrition and we looked at the microbiome, let's take a much broader vision. Let's look more holistically on human health and let's measure everything that we can about people. And the inspiration was the human genome project where we had the the view of the genome. And if we look at the 23 years that have passed since we first sequenced the human genome, that project taught us a lot. Basically, now we know hundreds of genetic variants that are involved in every single disease and and we reduced the cost of drug development by about 50%. But then when we looked back at that, we said that that's amazing, transformative, but it's only the human genome. It doesn't take into account any of the environmental factors, nothing about what we eat, how we exercise and where we live and and so on and so forth. And so we drew basically on the ability to execute these projects like we did and the human genome project to come up with now the human phenotype project, which really aims to eventually measure everything about people in as many time points as possible and track them longitudinally. And so to do that, um, we've that's what we've been doing for the past five years. We set up a clinic at my lab. Uh, people come in every day, dozens of people, and they undergo a very, very extensive set of tests, clinical, physiological tests that measure their cardiovascular health, that with imaging, we take high resolution image of the retina, for example, where you can see the intricate blood vessels, density, curvature, number and so on. And and that's really a window to the heart. You can with advanced AI tools really tell a lot about a person just by looking at that. Uh, we use DEXA to do full body scan and look at bone density, uh, ultrasound to look at fat in the liver and the carotids. Uh, we use sensors like the CGM, but also sleep sensors to assess your sleep quality for three different uh, nights.
Dr Rupy: Oh, they literally come in and they have a sleep study.
Professor Eran Segal: So so they do that from the comfort of their home actually, but uh, which is much more convenient, but um, but yeah, we have uh, these are sensors at the grade of a sleeping lab, except you don't go to a lab, you do it from home. And you know, we talk about nutrition, but of course, we know that sleep, sleep quality is really a major driver of of health. That really hasn't been studied. It's never been measured with these sensors like we're doing now on thousands of people. So having all the other information that we have like the diet of people and so on, uh, and their lifestyle, I believe and these are the algorithms we're trying to develop now, that we'll also be able to help and tailor your your sleep and basically be able to give you lifestyle behaviours that would improve your sleep quality. And again, sleep is just one of the aspects that that we're measuring. And then I think the other exciting aspect of the study is that we're looking at biomarkers like you mentioned. So we're looking really at the molecular level and together with all of this physiological and clinical information that I mentioned, we are also measuring the genetics, but then also microbiome, gut and vaginal microbiome. We're doing metabolomics, which is a study of thousands of different molecules in the blood. We quantify which ones are there and at what levels. We all know cholesterol, that's one of them, but we have thousands of of them. Many of these molecules, we don't even know what they are, but we can link them then to diseases. And once we find a molecule that has a strong link, we can go back and identify what that molecule is. Uh, we look at proteins in the blood, we look at RNA in the blood. Uh, we ourselves developed a proprietary assay for the immune system. So in one go, in one experiment, we can tell you all of the viruses you've ever been infected with.
Dr Rupy: Ever been exposed to?
Professor Eran Segal: Yes, at least those to which you currently have antibodies for.
Dr Rupy: Wow.
Professor Eran Segal: So COVID is of course one example, we can do that easily, but in that same experiment, we can identify your exposure to thousands of different viruses. And and this this really has meaning. So, uh, actually one of the, we never published this, but uh, one of the correlations we found was between Epstein virus infection, which many people are affected by, and a disease called multiple sclerosis. So this link between EBV and MS, we we found it. We didn't publish it because, you know, it was just an association. And then we saw last year, there was a beautiful study published by um, the database on the US military where they tracked soldiers for 25 years and they saw that EBV infection actually was predisposing people for development of of MS. So a lot of these associations can really have meaning and and again, this is just one virus and one disease. We have it for all the viruses and we can measure all of the different uh, disease outcomes.
Dr Rupy: EBV, my ears perked up then because a lot of functional medicine practitioners will specifically ask about a history of EBV infection. Um, and there is certainly loads of association, whether it's causal or not, who knows, but certainly there does appear to be a pattern there that warrants further investigation. And if you can get this bank of information, this these biomarkers, then we can begin to tease out who's going to be more at risk of previous EBV infection and autoimmune conditions like MS. Yeah.
Professor Eran Segal: So so that's really the vision to to be able to profile people very comprehensively. And I believe that this human phenotype project that we developed is is the most deeply phenotyped cohort that currently exists. And the vision is to track a very large number of such individuals. We profiled 12,000 so far, but our next milestone is to hit 100,000 and and do it not just in Israel, but also in additional countries. And and I think this is the type of data that will allow us to move from reactive medicine to preventative medicine where we'll be able to identify biomarkers that basically tell you where you are on the trajectory to developing disease or to staying healthy and and be able to uh, initially just tell you that trajectory and then at the second stage, also be able to intervene and move you back to give you advice that will move you back to the trajectory of staying healthy.
Dr Rupy: Yeah, I'm I'm a huge proponent of what gets measured gets managed. I think that's a common term that everyone's coming around to the idea of. And I think our current suite of tools, particularly in preventative cardiology, are limited in that, you know, there's still arguments raging around LDL cholesterol, apoB containing lipoproteins, all these different different um, uh, biomarkers that we have. But what I thought was quite interesting about your project is that you're trying to look for novel biomarkers that we wouldn't have even thought of and no one has even discovered yet. Talk to us a bit about how you're doing that within the field of acute myocardial infarctions and and preventative cardiology.
Professor Eran Segal: Yeah. Um, exactly. So, uh, so I think if you take any marker like LDL cholesterol, which people are arguing about, I think it again goes back to personalisation. That that's my hypothesis, that if you were able to overlay on top of that other biomarkers, maybe you'll find a stratification of the population in which for some individuals, high LDL cholesterol, maybe they go with some other biomarkers that are either protective or that um, that mitigate the effect of the high LDL cholesterol levels in in those individuals, but do not in other individuals where those biomarkers are are absent. And and and that stratification will uh, will allow you to get a more more a precise picture. So, so I think it's uh, it's not just measuring additional biomarkers, but it's really also doing the analysis to finding combinations. So I'll give you an example of one study that we did. This was on acute coronary syndrome patients, basically people who present with a heart attack at the at the clinic. And uh, what we found was uh, that we could actually take a personalised medicine approach to that. And what I mean is, uh, take even a single patient. Um, if you look at standard practice in general, uh, you know, your standard patient goes in, does some blood tests. As a general doctor, you look at the blood tests and you look at normal ranges for different um, of these blood tests and and you can say what's normal and what's abnormal. But that is uh, again, very general. Uh, of course, it should vary by by age, by gender, by So what we do here is we take a single patient, we go back to our very large healthy cohort from the human phenotype project, and then we identify a personalised matched cohort for that single patient. For example, a 60-year-old male who underwent a heart attack, we'd go to our healthy cohort, find 60-year-old males who did not undergo a heart attack, and then for every different marker, we would build the right reference panel. For many of these, there there are no reference panels because we are the first to do these measurements like all these different metabolites. So we build the right reference panel and that's a reference panel or a reference range of values that is personalised for that patient. And then we can see where that patient uh, falls and we know that we're looking at the right reference group for that patient. So when we do that, for example, for the thousands of different metabolites, then for each patient, we can identify those metabolites, those blood tests that are abnormal for that patient when compared to his correct personalised reference matched group. And then we can look at all of those deviations and we can ask, are these deviations metabolites that are affected by diet or by genetics or microbiome or lifestyle factors and so on. And by that, identify the pathophysiology of disease for that patient and also what should be targeted for that patient. Because if I have a patient where I can see that the metabolites that are disrupted are metabolites that I know are generated by good bacteria, then it's not going to help to um, uh, tailor the diet for that individual. It's going to help to go and do a maybe a a targeted probiotic intervention, some intervention in the gut bacteria and and vice versa. And we know today that on average, by a rule of thumb, every medication that you give is only effective in about 20% of the patients that you give it to them.
Dr Rupy: 20%?
Professor Eran Segal: I mean, this is just a, you know, a ballpark number, of course, it varies by by different drugs and indications, but yeah, on average. And uh, you know, all the amazing developments in cancer immunotherapy, for example, um, we have some cancers like lung cancer, which was incurable several years ago and you'd know you you'd die within a few years. Now, 20% of patients who receive cancer immunotherapy for lung cancer, they fully recover from it. But then the other 80%, for many of them, it uh, goes back within a few years. And for some, it goes back um, even faster and and eventually it's not uh, it's ineffective. Um, and there, for example, people have been studying also the gut bacteria and they've been finding that, for example, sometimes there are bacteria in the tumor itself, which are breaking down the immunotherapy and rendering it uh, ineffective. And so the thought and a lot of uh, even pharma companies and researchers are now working in these directions is that if you give the cancer immunotherapy together with a treatment and a targeting for those bacteria, you may be able to render the cancer immunotherapy to be effective.
Dr Rupy: Fascinating.
Professor Eran Segal: So again, this is just one example where we understand some of the mechanism perhaps, but but again, as a rule of thumb, um, medications that are uh, generally effective on average, again, statistically, like we talked about the glycemic index, um, you want to be one of the lucky ones for which the medication works, but uh, for many it doesn't work. And I think the reason is because the underlying root of the disease varies between people. This is what we're seeing with acute coronary syndrome patients, and this is likely going to be true for most diseases.
Dr Rupy: Just to zoom into that ACS patient, the acute coronary syndrome patient a bit more. I think you articulated this really well in your lecture about how the disease risks are going to be different depending on the the patient that you're seeing. In some cases, you know, more of a genetic risk factor is coming into play. In other cases, they can have a great diet, um, but actually it's, you know, the environment in which they they live in or stress or whatever it might be. So you're going to be able to give us, hopefully, a clearer picture as to which determinant, which risk factor is having more of a of an of an impact on someone's ultimate uh, disease.
Professor Eran Segal: Exactly. Yeah. So, so again, I think the idea of profiling individuals and then comparing them to really what is what is the right reference panel specific for them will allow us to highlight for people and not just for acute coronary syndrome patients, but for everybody, to highlight uh, where they are in terms of um, overall health and which body system uh, perhaps is affected. This relates to other studies that we're doing on just aging of different body systems. And um, and yeah, and so and so by that, we can identify uh, where you have abnormalities and what should be targeted in different people. And we know that uh, if if your diet is fine, then, you know, targeting that is not going to help you.
Dr Rupy: That's super fascinating. And in terms of the uh, use of probiotics, something that we get asked about a lot. There's lots of probiotics on on sale, online with various studies supporting their efficacy. If I'm taking anything away from what you're saying, it really depends on the patient cohort that's going to respond positively or negatively to the probiotic that you're introducing, beyond just the strains that you're introducing into the digestive tract as well. I'm assuming you've done a number of different studies looking at various strains in using your cohort.
Professor Eran Segal: Yeah. Um, exactly. So we've done uh, studies on some of the available probiotics today and I'll tell you what we found. We found that uh, first of all, uh, colonisation uh, really varies between people. So we took, you know, one probiotic product which has 11 different bacterial strains. We saw that in many people, there was no colonisation. So just the probiotics, they go in and then they go out the same way, exactly. And in others, there was colonisation of just uh, a small number and it varied really between people. Uh, and so to the extent that part of the effect should be mediated by colonisation of the good bacteria, for many people, and this is again probably in a personalised manner, for many people, it's just not going to have an effect. So that's so that's one discovery we had. Uh, the second was actually, I think very interesting. Uh, you know, um, doctors prescribe probiotics, uh, you know when? Uh, after you give antibiotics for whatever reason you gave them, you people you typically would uh, this is general practice, you tell people, okay, now you should take probiotics in order to restore your uh, gut bacteria. So what we found is that if you give probiotics after antibiotics, then in contrast to the lack of colonisation we saw before, after antibiotics, we see much more extensive colonisation because the antibiotics basically cleared the niche, allowing for probiotics to now thrive. But then what we discovered is that that colonisation of the probiotics actually delays significantly restoration of your original microbiome. So if you had a healthy and good microbiome and you know, now you got infected by some bacteria, you had to use antibiotics and of course, you should use antibiotics, then taking probiotics after that would delay significantly by even months restoration of your original microbiome. So this general advice that we're giving people has many uh, has many issues that we weren't aware before of before the studies that that we did. This was actually part of an episode of 60 minutes in the US. We we went on on that show to talk about these discoveries.
Dr Rupy: Wow, that's brilliant. We'll link to that in the show notes for sure. So that's a case of where the probiotics are doing a great job of colonising, but the net effect is actually something that you don't want.
Professor Eran Segal: Probably probably negative, exactly. So we didn't study the long-term clinical impact of not restoring your original microbiome, but uh, I'm just saying this is an observation that uh, yes, it may have bad clinical effects, we just don't know. And so just blindly taking antibiotics or prescribing them after uh, antibiotics may actually have a negative impact. Those who are taking probiotics where it's just going in and going out, it's probably not doing any harm except um, you know, it's costing you money.
Dr Rupy: Yeah, yeah. Is there ever a situation where you've seen some of the cohort introduce probiotics and it have a great colonisation effect? Are there like hyper responders to probiotics that you've seen?
Professor Eran Segal: I think the studies that we have done so far have been on too few people to be able to to say that. But uh, again, I just want to emphasise that I think the concept of probiotics is actually a good concept because uh, because our studies also show that some of the bacteria that we have just our some of our commensal bacteria are probably not good for us. Some of them are giving for some people a propensity for obesity or for developing diabetes. And sometimes we're missing some protective bacteria. For example, in the acute coronary syndrome patient study, we actually identified bacteria that we think are protective that are absent from ACS patients. And so a follow-up study that we want to do is now to actually give these specific bacteria to uh, high risk individuals, individuals who for many different factors are at risk for cardiovascular disease and lack this bacteria that we identified. That's a probiotic if you will that we want to give them, but but you see it's it's um, it's a data-driven one where we first analysed a very large cohort, identified which bacteria in a data-driven manner are either missing or shouldn't be there that we'd like to either introduce or replace. And then we go in surgically with the right bacteria uh, in order to give them and see the the benefits. So so I think this is the future, this is what we should be doing in terms of the probiotics and not just the ones that we're giving today.
Dr Rupy: Yeah, it's almost like you're giving drugs in a targeted manner, but in this case it's probiotics. Do you remember the strain of those particular bacteria that you might be looking to introduce into someone's?
Professor Eran Segal: Yeah, there's there's a few of those and I can I'll send you a link to this was published in Nature Medicine a couple of years ago.
Dr Rupy: Because the one that everyone always remembers is the one that introduces weight loss, Akkermansia.
Professor Eran Segal: Yes. Which by the way, uh, I'll say that now that we have our studies on 12,000 people, by far the largest microbiome court, we're actually finding that Akkermansia is not really one that yeah. And we're finding many others that that are. We're actually we actually found that Akkermansia is actually providing support for other bacteria that do have some benefits. So by itself, according to our data, if you partition people between those who have Akkermansia and those who do not, we don't see a difference in uh, weight. But for many other bacteria, which are the ones we want to give as probiotics, we do see. And for some of them, they're correlated with Akkermansia. So we actually think it's a secondary finding and not really the the root cause. And I think this shows you uh, what we study and learn from small numbers versus what we tease out and identify robustly when we go to uh, larger numbers.
Dr Rupy: Yeah, it's almost like the deeper you go, the more complex it is and you realise actually it's collections of different strains rather than the singular strains uh, working alone.
Professor Eran Segal: And it's also the power of numbers. So, um, something that we published uh, was uh, you know, there there's if you look at microbiome studies, which are typically done on still on dozens, sometimes a few hundreds of people, those are small studies. So if you look at many of them, they'll be uh, contradictory. So some bacteria, you know, like Akkermansia would be found in obesity in one paper, but not found in another paper. So so how do you reconcile all of these um, contradictory findings? So what we found is that if you take our very large cohort and you sample dozens or hundreds of people and you ask, okay, which bacteria are relevant for obesity, you'll find one answer. You sample a different set of individuals, you'll find a different answer. You'll sample a different, you'll find a different answer. So just by chance, you'll sample a small number and you'll find some association, but those will not be robust and you won't find them on uh, you know, 10,000 people. So it's only when you start to go to larger numbers that you do the sampling again and again and again and you get the same answer. So you really need, you know, if you if you're doing statistics, you need to do them with uh, large numbers and and hence the human phenotype project which which really aims to go to hundreds of thousands of individuals.
Dr Rupy: In terms of the next milestone, it sounds like it's 100,000. In reality, if we're really going to get precise about genomics and and uh, uh, dietary advice that's truly personalised, introducing other biomarkers and metabolomics, etc, all the omics, how many people do you fathom you'd need to have in order to get to that level?
Professor Eran Segal: So, I think the more that uh, we'll get, we'll be able to get more and more accurate results. Um, I think with with the 12,000 that we already have, we already have been able to make a lot of discoveries. I think we'll be able to make another major leap with 100,000. And then once we get to, hopefully we get there and then um, I don't know if it'll be me, but maybe somebody else will uh, take the next leap to to a million. I think we'll be able to make more advances, but then um, they'll be for conditions that are probably more and more rare. So not to say that it's not important, but um, I think the most of the common factors, common morbidities that that we're seeing, I think a lot of them will be able to uh, study and identify with 100,000.
Dr Rupy: Let's talk about faecal transplants because you mentioned this uh, in your lecture as well and I've I've seen you mention it in previous talks. Uh, and there is a bit of an ick factor, I think. People are coming around to this idea, but uh, what I thought was interesting is actually you're using the donor's own matter. So what kind of things have we have you gleaned from the the studies you've done thus far and in which particular areas are you concentrating?
Professor Eran Segal: So we've done this uh, for atopic dermatitis, a disease of the skin, a very prevalent disease in about 20% of children, 10% of adults to various degrees. Uh, we did initially a proof of principle study just on nine patients, but all of them responded amazingly and positively with reductions of clinical manifestation of disease within two weeks of providing them with bacteria with with faecal material from a healthy donor. Um, we're now doing a larger study. Um, that's what we've been studying. Of course, uh, FMTs are uh, for a infection in in the hospitals for Clostridium difficile infection, um, they're really now uh, they have amazing efficacy and and they're they're being used. Um, but I think eventually this is not the approach that we'd like to take, uh, because um, it it's very hard, it's it's not it's less reproducible. The material that you'll be giving will vary even within the same person and of course, across different individuals. Uh, and it's also a very crude uh, intervention. And so, uh, really the uh, the the way we're going is to try and do it much more surgical. So to identify the specific bacteria that are the ones providing the therapeutic benefit, extracting them, growing them in a GMP facility, uh, and then providing them as a data-driven probiotics, if you will, uh, to patients and and seeing if that can be beneficial. And I believe that by choosing the right bacteria, we can have an impact on virtually every human condition. There there's even FMT studies not that we did that others did on autism. So you know, in autism there's also an association with various GI tract issues. And uh, it's been shown even in children to increase their overall function for autistic children. Again, still small scale studies, but just showing what I think is the potential. Uh, and but eventually, I think also there, FMT would not be the way to go, but we really should try and understand and identify the mechanism, the specific bacteria that we'll put together as hopefully a therapy also for autism, but also for many different other conditions.
Dr Rupy: Yeah, because I was going to ask about that proof of concept study with the atopic dermatitis. How did you select the healthy donors and how did you convince yourselves that, okay, this healthy donor with an absence of atopic dermatitis is going to be beneficial even of itself for the person afflicted with the condition?
Professor Eran Segal: Yeah. So, so we didn't know. I mean, we we just we just tried and um, and and we basically took uh, young healthy donors who are overall following what we think is a healthy diet and you know, living a healthy lifestyle. Uh, but we didn't know and actually in the follow-up study that we're doing now on a larger population, we have a donor where um, we didn't see big uh, big effects. We switched to a different donor and now we are seeing big beneficial effects. So so there's no doubt that the donor is uh, is going to matter. Uh, the studies that were done so far are still at too small of a scale, both our studies and in general, FMT studies because because they're also hard to execute. Uh, so, uh, but I think so so here too, the power of numbers that we talked about before, I think that will also be needed before we tease out and identify the specific bacteria that uh, that we need to provide.
Dr Rupy: On that note around being more surgical about probiotics, uh, everyone seems to focus on the bacteria that you find in the in the digestive tract. But down there's not just bacteria, right? You have nematodes, fungi, viruses. How can we determine that it's not a collection of other microbes that might have a supportive role for the bacteria that aren't also enabling that effect that you see?
Professor Eran Segal: That that's a great question actually and uh, and we don't know that and until we actually go in with the specific bacteria and compare it to the FMT, we won't know the answer and and you're right, it but it it's also not just the other viruses and fungi, it's also when you take a faecal matter, it contains uh, other proteins and uh, other um, elements that are and metabolites that are secreted by these uh, different uh, living creatures that that we have. Uh, so it may be part of uh, the effect of those. So, so actually that I think that's a great scientific question which we are and everybody is aware of and we won't know until we actually go in surgically with some bacteria. Um, so again, the approach that we're taking is because it's not realistic to um, do an FMT from hundreds of people to tens of thousands of recipients, we're taking um, you know, a data-driven approach by analysing our population of 12,000 people to identify there from the very large scale data, which bacteria are relevant for which indication. And then hopefully by finding those robust findings and going with those bacteria in, those will actually uh, that will increase the chance that these bacteria will have an effect.
Dr Rupy: So I think that's our best shot on goal for that.
Professor Eran Segal: Yeah, no, I agree. I think taking that data-driven approach using the tools and the knowledge that we have today is is almost like the next point of um, uh, the proof of concept step to take when it comes to these studies. And I I guess like if we sort of fast forward five, maybe 10 years in the future and you have these collection of metabolites from all these different cohorts and we have a lot more robust understanding about these associations and how much diet, genomics and all the other risk factors play, how do you see the future of of medicine? You know, do people need to see a a doctor or need to see the surgeon? Can we can we really envisage a world where everything is essentially individualised and personalised from your diet, your exercise routine, your supplement routine, your herbal medicine routine, if that's included in that that smorgasbord of of treatments. What is your sort of vision in that respect?
Professor Eran Segal: Yeah, so I actually think we'll we'll never and we don't want to take doctors out of the equation. I think we want to just empower doctors with much more sophisticated tools that really provide them with more information so that they're not in the dark so that they don't just prescribe probiotics after antibiotics without knowing what effect that will have, right? Like we talked about before. So I think it's really about empowering doctors on the one hand, but also empowering patients and just just people with the ability to have tools to track their health on a continuous basis. Again, not uh, you know, not obsessively, not all the time and of course enjoy life and so on. But but I think um, I think we can move a lot of the testing that we're now doing on occasion in the clinic and many people are not compliant with, we can move that uh, to the comfort of our home, um, using technologies, using the smartphones that we all carry, using smart watches, using other sensors. Um, and then when there are issues, we would go to uh, we would then go to an educated person, we would go to a doctor, but that doctor would also understand what we have been measuring, would know how that all links up with previous findings and would be able to give us better advice empowered by all these data and discoveries that have been made.
Dr Rupy: You've got a fancy lab with lots of all the tools that everyone could want, you know, as a computational biologist. If there were a selection of biomarkers that people should get regularly tested of all ages and you've only got five, a maximum of five, um, which ones do you think we should be leaning into more? I can almost guess the first one, but I'm going to let you I'll I'll let you and you're limited in terms of the the complexity. It needs to be those kind of biomarkers that somebody could get regularly available either sent to them or or generated by their doctor.
Professor Eran Segal: Well, I mean, you know, I think we can't exclude uh, you know, a hundred years of of medicine which really identified uh, a lot of the biomarkers that are really key. So, yes, I think if we are to track five biomarkers, yeah, I would track the the ones related to I would choose one related to to glucose management, of course. Yeah, so haemoglobin A1C or fasting glucose or using a CGM. Then some that are related to uh, to lipids in our blood. Um, yeah, so so a lot of the standard ones if we just had five and and I think we identified those that give the greatest uh, value for money, but again, that's just the tip of the iceberg and I think once we identify now many different um, biomarkers that are relevant, then we'll be able to also personalise them and and for different individuals, it'll be different markers that we want to follow because, you know, haemoglobin A1C is very key because we want to control our glucose, but for many people, like those lucky people who almost eat everything and never spike their blood sugar levels, you know, for them, it's it's not a relevant biomarker. Right? So for them it'll be something else. So again, we're going back to statistics. If you had to choose five for all the population, you'd probably choose the ones that we have been following. But uh, if you're able to tailor them for people, I think you would make individualised choices on even which biomarkers to follow for which people because some people from their profile, you would know are more prone to diabetes, some people are more prone to developing neurodegenerative diseases and some people are more prone to developing cardiovascular related diseases. And the biomarkers are different for each.
Dr Rupy: Are you have you made any changes yourself since doing this over the last how many years?
Professor Eran Segal: Absolutely.
Dr Rupy: What are the key things that you've changed?
Professor Eran Segal: So this whole thing actually my whole getting into this actually started from personal interest in running.
Dr Rupy: In running?
Professor Eran Segal: Yeah, so I'm an amateur marathon runner. I I had a dream to break the three hours in a marathon. And I, you know, continuously improved my running skills, but then I realised that diet is important. Um, but back then I was eating just following the general dietary advice, you know, we do carbohydrate loading the night before you go for a long run. After you you uh, you do the long run, I was very hungry, so I also ate also carbohydrates to restore all of those that I depleted during the run. And then realising uh, by measurements that a lot of these carbohydrates were spiking my blood sugar levels, I just stopped eating them before runs and I stopped doing carbohydrate loading and now I could go for a 30 kilometre run and eat just a regular whatever salad and egg or whatever meal the night before. Not eat anything before I go on a run and surprisingly, actually not even feel hungry, feel very energised during the run and not even feel hungry hours after I complete the run.
Dr Rupy: Wow. Yeah, yeah. What do you think you're tapping into? Do you go into ketosis or do you you tap into fat stores?
Professor Eran Segal: So I don't think I personally am in ketosis because I don't uh, almost completely avoid carbohydrates, but I think I tap into for sure much more um, fat utilisation for energy uh, during exercise, uh, which I think at least for me has probably a lot of uh, benefits. Um, and I think the other thing, this is this is not uh, not scientific, but more of a hypothesis. You know, I think the reason why you may have a difference between feeling hungry or not, if you think about your carbohydrates storage, an average person would have about 3,000 calories that you can store in glycogen for carbohydrates, even if you do carbohydrate loading. 3,000 calories is is about the amount that you need for a marathon or if you go on a 30 kilometre run for the average person, you'll deplete most of that. So if you think about your the sensors that you have in a body, you take a system that's uh, after you carbohydrate load that's at 100%, you drive it down to nearly zero after a run, your body senses that and signals your brain that you're now low on carbohydrates, you should upload and uh, uptake more carbohydrates, right? You have these sensors which trigger hunger. And I think that if you tap into fat for energy, you know, an average person stores about 60,000 calories of fat in your body. This is why we can fast for for many, many weeks and uh, and still survive. Uh, and so if you go even for a long run and you go down from 60,000 to 57,000, um, you know, you've only depleted 3 or 5% of that and maybe then your sensors uh, you know, are not signalling that you're you're hungry. This is just a hypothesis, but um, but there is a very intriguing difference. For me, it was very surprising that um, going on long runs without any carb loading, um, and then not feeling hungry, I could I could go almost a full day after and not eat. Um, was very surprising.
Dr Rupy: Well, this is this is what's interesting about your work because I think it's going to explain those differences that people anecdotally have, the fact that some people can fast in the morning and train on that and do their some of their best gym work or their PB runs versus other people who are like, well, I can't I can't uh, run without having a proper carbohydrate load and you know, they need to have all that their their food and stuff before that. So it is really interesting to to see that. And one of the other things I was going to ask about was uh, and I've I've literally just lost my um, train of thought now. Uh, supplements, that's it. In terms of all the information that you've gleaned from your own um, information, your own biomarkers, have you have you changed your supplement routine or are you bullish on supplements or is that something that
Professor Eran Segal: Actually I I right now I'm not taking any supplements. I was vegetarian for a while and I took B12. Now I went back to eating meat and I stopped taking B12. Um, yeah, in general, um, I believe that if you're eating the proper diet and you're eating a diverse enough diet, then and you're healthy, of course, you're not lacking uh, something in particular, then you should track yourself, but in most cases, I think you'll be able to maintain balanced, healthy levels of uh, all of the different factors without taking in supplements. I think I think that should be the goal.
Dr Rupy: Yeah.
Professor Eran Segal: To basically get things naturally.
Dr Rupy: That should be the goal. And I think because we're we're using tools that essentially give us the reference ranges for the average, if my vitamin D3 level, for example, is low, I'm inclined to push that a little bit higher because of what I know about vitamin D3. Putting aside the fact that I don't have clinical symptoms of vitamin D deficiency, I just want to make sure that I'm doing the right thing and I'm topping myself up appropriately. But, you know, outside of that, I think a lot of people are finding themselves in the same conundrum to supplement or not to supplement. So, I mean, I err on the side of, okay, I'll supplement a little bit, but I get your point. I think the goal is to not have to supplement if possible.
Professor Eran Segal: And you know, it's there's also a difference between if you supplement whatever, vitamin B12 and your levels of vitamin B12 in the blood now became normal, it still doesn't mean that, you know, then there's absorption and is your cell, are your cells really able to utilise that vitamin B12 and and so on. So, so even taking the supplement and even seeing that it has an effect is is not telling you the full story. So, you know, I think in the end, um, you know, when when you eat when you eat an apple, yes, you are also eating some glucose, it may spike your blood sugar levels a bit, maybe, maybe not, but you're also getting all of the other nutrients that are in the apple. Many of which, by the way, we also don't even know yet what they are. But um, you know, they were conducive for that apple to grow. So, uh, I think by that analogy, eating whole foods, eating less processed foods that every processing that we do to the food, it alters it, takes away some ingredients that were relevant for its growth and maybe introduces some other ingredients during the process that are not good. So, you know, I think eating naturally, healthy, whole foods that are right for you, I think that would be what we should strive for. That's what I strive for.
Dr Rupy: Yeah, yeah. Well, you're a shining example of that. Um, you mentioned that you're looking for collaborators across the world. We actually had Professor Kuner on the podcast, um, who does the South Asian uh, biobank study, um, as a compliment to the UK Biobank because South Asians are underrepresented. We have lots of risks of type two diabetes and and the like. Um, what kind of collaborators are you looking for and in which geographical regions?
Professor Eran Segal: Yeah, so, uh, we are looking to expand the human phenotype project as I mentioned. Uh, we're really looking for expanding it not just by the number of people, but also culturally, ethnically, genetically. Um, so, yeah, any population, I think the UK has a diverse population, much like we have in Israel, but a different type of diversity. So, uh, here would be a great place to open up a site and open up a clinic that would uh, do all this very extensive profiling. And I think we'll learn different things. And when we'll combine it with the other data sets that we have, we'll be able to also arrive at much more robust findings that are not just specific to uh, to the Israeli population that we've been studying so far. So, if anybody is hearing this and is interested, uh, we basically have a platform where we can very efficiently at relatively very low cost, um, speed up uh, profiling of individuals with everything that we have been doing or even a subset of that. Um, and so people who are interested in following up such cohorts, I invite them to uh, to talk to me.
Dr Rupy: Yeah, epic. Well, you're doing incredible work and I'd love to support in any way possible, but appreciate your time, man. You're awesome.
Professor Eran Segal: Thank you very much. Thanks for having me.
