#173 – Biomarkers: Observing oncoming sickness patterns & measuring good health | Andrew Herr & Azure Grant

Episode introduction

Episode Transcript

Andrew Herr (00:06):

If you can combine running experiments with high fidelity, high frequency data, that to me opens up the future of the field. And so, I think where Levels has made high fidelity, high frequency data available to more people on this critical metabolic parameter on blood sugar, that’s one step to being able to run and better interpret experiments. But as you go from a few things you can measure like that to tens of things, then the covariate and other factors are important. But our big thing is you’ve got to be running experiments, otherwise you don’t know what caused what.

Ben Grynol (00:52):

I’m Ben Grynoll part of the early startup team here at Levels. We’re building tech that helps people to understand their metabolic health. And this is your front row seat to everything we do. This is a whole new level.

Ben Grynol (01:17):

We’ve been talking a lot about biomarkers, why they are important, and why it’s integral to understand all these different markers in real time. What can we glean from having different insight about biomarkers and how they fluctuate. How they oscillate in her body in real time all the time. And so, Andrew Herr CEO of Fount and Azure Grant, part of our research team sat down and discussed the importance of network physiology, if you want to call it that. Our body is an ecosystem in itself of all these interconnected biomarkers that are in constant fluctuation depending on lifestyle factors like food, diet, sleep, exercise, stress, environmental factors, the list goes on. So, the two of them sat down and discussed all these different concepts around biomarkers. What to mark? When to market? Why we should market? And how we can think about it moving forward. No need to wait. Here’s Azure with a deeper intro on Andrew.

Dr. Azure Grant (02:16):

Hello Everybody. I’m Azure Grant research design lead at Levels here today with Andrew Herr, the founder and CEO of Fount. Which is a concierge service that combines wearables, blood testing, and expert feedback to help people improve their health and performance. So, today you’re going to get to know Andrew and me a little bit better and we’re going to focus on three main questions, they’re actually quite big questions. Our frameworks for understanding when to measure and how to interpret biological time series, what markers are measured in Fount and really interesting. And then, a fun one, what are our dream lists of biomarkers to measure in the future? So first we’re going to introduce ourselves. Andrew has a master’s degree in many things, microbiology, and immunology, security studies, and health physics actually had to look that up. Study the interaction of physiology and radiation at Georgetown University.

Dr. Azure Grant (03:11):

And he also completed undergraduate work there in the school of support and service. And among his many achievements, he’s been awarded the title of Mad Scientist by the US Army, which was not even a title that I was aware they awarded to ED1. So today we’re going to draw on all of that experience to talk about a lot of blood biomarkers ought to be measuring and why. So Andrew.

Andrew Herr (03:31):

Thanks so much, Azure. It’s great to be here and looking forward to sharing with the audience on some of the things we measure, why we measure, and what we want to measure.

Dr. Azure Grant (03:40):

Awesome. And just a little bit of short background on me, I’m doing research design at Levels, meaning putting together new studies and analyses of our retrospective data and with the community. My PhD background is neuroscience from Cal Barkley and I worked a lot on biological rhythms and neuroendocrinology. So, as we’ll probably be apparent, Andrew as we talk. I’m super interested in how metabolism and the reproductive system interact over life and contribute to overall health. So, first Andrew, could you say a little bit more about your story. How you decided to study what you study, how you developed an interest in improving health, especially at the very high end of the performance spectrum?

Andrew Herr (04:22):

Yeah, I would say that if I lived my life 10 times, I would’ve been a doctor five of them. My dad’s a doc and my mom’s side of the family is a bunch of marines including some real heroism and things that are really proud of. And so, I came out with this dual interest in the science side, in the history national security side. And I was lucky to find a program where I could do a degree in science technology and national security at Georgetown. I did that work, decided that if I really wanted to work at the intersection of science and technology with these really important applications, probably wanted to go really deep into the science. I was working on nuclear and biological weapons issues and probably don’t want to get those wrong technically. And so, I ended up doing the graduate work you described in biology, physics, their interaction.

Andrew Herr (05:10):

And one day I was walking around campus, I got a call from the career center and they said somebody walked in, they said they need someone who does biology, national security and can write. And they’re like, “Well we know this guy, Andrew.” And I got hired by the secretary defense’s personal think tank on the future of national security to work to run their portfolio on the future of biotech. And so, it was a really amazing opportunity to think about where every element from performance optimization, from how understanding physiology, what’s the physiology of leadership under stress and how would you leverage that to build better units all the way to threats in this space that range from the biological weapon side to other let’s say unpleasant discussions. So, I worked that and then over the next seven years I ran various human performance and biotech strategy efforts, from hoping to write their research and development strategies down to getting Navy Seals ready to deploy, ran a big clinical trial.

Andrew Herr (06:11):

So really was amazing to have an opportunity to run every element of human performance. And then, out of that I’d be speaking at conferences and executives would come to me and say, “Hey, I get what the Navy Seals, but what should I be doing?” And so, that was the seed of what later became Fount developed, really one on one individualized programs for executives. Really focused on the idea of can we run experiments to figure out what’s optimal for your body, your goals, and your lifestyle. And so, we call this experiment journeys, this is the methodology we call it. And then, have now built really amazing technology and team around doing that. And that’s where we are at Fount.

Dr. Azure Grant (06:52):

That is super cool. I didn’t realize that there was such a personal experiment component to what you guys are doing and I would love to hear more about that. It’s something that I would really like to see more of at Levels as well. And I’ve interacted with this mostly through the quantified self community, which are more everyday people are often people with undiagnosed disorders who are trying to learn the scientific method and apply it in their own lives, and how to scale this to more people, or lower the barrier to entry, or especially scale it to busy people, is always one of the hardest things. So, I would love to hear in a little while how you guys approach that problem. But first I wanted to do a little bit of framing, not even about what to measure, a little bit about when to measure.

Dr. Azure Grant (07:43):

So these days of course for our listeners, we have more and more opportunities to measure biomarkers as frequently as every few minutes or if the measurement is taken from something like a skin sensor even faster. So if you think about things like heart rate variability, breathing right, body temperature can get those very, very quickly. Then we have things like blood glucose that you can measure every handful of minutes as a first blood biomarker going continuous. So this enables the device to not only capture a single blink in time out of context, but also to evaluate a variable as a person goes throughout their day, takes actions, has experiences, and wants to relate those experiences and actions to what’s going on inside their body. And it also allows us as the researchers to tease out patterns of change over time that could maybe characterize a particular illness or a particular sign of good health and performance. So today, given that context that we’re surrounded by a combination of outputs that can be measured 24/7 and some that are still limited to single time point blood draws, what is your framework?

Andrew Herr (08:51):

The way we think about it is really dependent on the client’s goals. So when we step back, we think about helping clients optimize to their body, their goals, and their lifestyle. And so, if your goals are about energy and focus, those are largely subjective quantities. And so, there we’re going to ask you directly about it, obviously you could actually measure objective focus, but it’s really like when do you want to focus? So some people for example will say, “Hey, I have these three hours in the afternoon, that’s my deep work time, I want to drop into focus then.” And so, our goal is to measure when you want to be in that state, how easy is it to get in? How easy is it to stay in? And if you pop back out, how easy is it to drop back in? So, I think the answer is I use that to say there we might be wanting to do that daily and daily might be as high a frequency as we need.

Andrew Herr (09:47):

Whereas if you’re running experiments with nutrition, for example with metabolic effects, so for example levels in amazing blood glucose monitoring. Then, you really want to see that semi continuous curve, their five minute sampling is fine, it doesn’t matter if you know the microsecond. And then, when we go from there, we think about our pro athletes or elite athletes and there it actually may matter to measure things at that very fine, almost truly continuous level. If you’re thinking about reaction time, which doesn’t matter that much for a CEO usually, but is very important for Navy seal in certain circumstances there you can’t measure every five minutes because in between those five minute gaps may be the critical moment. So, I think it’s really dependent on the person, their goals, their lifestyle. But at the end of the day, higher frequency, assuming you don’t lose fidelity is almost always better, as long as it isn’t too inconvenient.

Andrew Herr (10:50):

I don’t know anyone who wants to go to the lab twice a day every day to get real venipuncture blood. And I guess the last thing I’ll say is, the more we know about the interventions and the individual, the lower frequency we absolutely need. So, if we know with reasonably high fidelity that a supplement or an exercise protocol or nutrition is likely to move blood markers in a certain direction, then it’s slightly less important to have it at high frequency because we can interpret. So we typically run one to four experiments per week with our clients and we’re picking from of nutrition, supplements, meditation, sleep, light, temperature, exercise, recovery. And if we give you between your monthly blood draws, if you’re doing let’s say 10 experiments and we don’t know, some are pushing things up, some are pushing things down, then that really might not be that helpful.

Andrew Herr (11:47):

But if we know with reasonably high fidelity what’s likely pushing what direction, and we have other markers that inform us in between, typically if your inflammation’s coming down, you’re going to feel differently, it’s going to affect your sleep differently and we get sleep every night. So again, we can use some of these markers to infer and inform what the actual measurements mean, even if some of them are lower frequency because of convenience or as this rolls out to more and more people costs.

Dr. Azure Grant (12:18):

Gotcha. That makes a lot of sense. And what I’m hearing from you, correct me if I’m wrong, is that you’re working really closely with people to understand their subjective experience and needs on a more frequent basis, when something continuous might not be available to supplement in between those blood draws. Is there anything that you found really helpful in terms of how to capture subjective experience? Is it mostly good app interface design to you or frequent interviews with a coach? Little off topic, but I feel like it’s such an important problem that pretty much the company that I’ve talked to really wants to understand.

Andrew Herr (12:56):

What we’ve found is that you can, by asking the right questions and explaining to people why you’re asking questions, help train people, for lack of a better word, to understand their body. I mean we all have this insanely powerful wearable that’s hardwired into our body up here in our brain. And so.

Dr. Azure Grant (13:16):


Andrew Herr (13:17):

It’s fully biocompatible as best I can tell. So, we have this brain, it’s hardwired into the system, it’s getting all these feedback, but often people don’t recognize or realize what the patterns mean. So, after certain meals, if people find their energy dips and their anxiety goes up, that’s a classic response to inflammation in the body, in the brain. And so, once they realize that though, they see it. And so, at the beginning of the program we do a 90 minute interview with coach where we really go in depth on goals, day to day schedule, and then deep dive in everything from the performance variables like energy, focused, mood, athletic performance, all the way to family history and other factors. And through that hour and a half there’s almost never a case where the client doesn’t have an aha moment. And then, at the beginning of the program you’re checking in with your coach weekly and so they’re having a discussion and those questions are coming back up.

Andrew Herr (14:12):

And then when you combine that with app design, that’ll makes it easy to answer questions more frequently. And obviously the ability to see what those questions mean and visualize data. It’s like train people, make sure you’re asking the right questions that actually cue people to the data you want, and then make it easy for them to give you that data. And I think that three step process really delivers incredible subjective data. And then we combine that with the objective monthly blood work, sleep data every night, and haven’t announced this yet publicly, but looking forward to some more partnerships, especially with you guys at CGM space, as well as things like full body MRI for cancer screen, et cetera.

Dr. Azure Grant (14:58):

Awesome. I totally agree. I mean one of the things that I’ve learned so far is even with minute by minute data can work really hard to pull out features that are predictive of something. So for instance, ovulation prediction and confirmation, was a really fun thing to work on. And, often along the way when I would work with individual people and point out the signals either in their glucose data, so blood glucose often goes up after ovulation, insulin sensitivity is a little worse, things like heart rate and body temperature can have a little dip and do some interesting patterns before ovulation. People would start to be able to intercept these things a little bit better as they went along. So I hope that’s an end state.

Andrew Herr (15:44):

Yeah, I would say what are the physiological processes that, for example, on the physical side of PMS cause discomfort and can you mitigate those? And we’re really excited to launch a product probably later this year, early next year where we have about a 90% rate of being able to mitigate the physical symptoms thereof because so many of them are inflammatory glucose sensitivity, things that are modifiable. But, if you match the right interventions to the right part of the rhythm. And so, there I think is exactly your point. Either you need to know where you are in the rhythm objectively or the person needs to be able to infer it from their subjective state. And that’s what’s really cool, as we think about pairing interventions with diagnostics.

Dr. Azure Grant (16:29):

Okay, I have about 10 follow up questions that I want to ask you on that. And I’m going to hold myself not to until we get to some of these biomarkers. But I think what you’re saying tailors very well with the framework that I’ve learned about so far and that I’ve seen at Levels. Which is that you really don’t have a choice all the time about how often you want to measure and often you’ll be kept to infrequent measurements, but when you can go continuous, the patterns in that data can predict the features which are really interesting. So, because substances in our body vary rhythmically and therefore fairly predictably over time, like the waves of the ocean, you can use the circadian or daily rhythms in those signals. But also faster things like every few hour rhythms, or slower things like ovulatory rhythms in women, or seasonal rhythms in everyone. To both put a marker in temporal context and be able to interpret it better, and to do fancy signal processing and feature detection to create more specific biomarkers.

Andrew Herr (17:33):

We think a lot about the customization individualization piece. And right now the data does not… It’s not possible to do really deep customization just from blood work, or wearable data, or others. There’s just not enough understanding of the interaction between cause and effect there. But as we run more interventions and you see the differences in response and then per your point you have these patterns beforehand that you could then use to predict who’s going to be category A, B, C, D, E, F, G responder. I think the continuous thing gets more and more valuable over time.

Dr. Azure Grant (18:10):

Yeah, that’s one of my favorite things. I’m curious what your types are because I’ve started to think a lot about types by age and in particular reproductive age. So, for women, are you on or off birth control? Are you trying to have a kid and maybe pregnant? Are you entering the perimenopausal years? For guys, it’s often where are you on the spectrum of youth or heading towards and andropause or have you recently become a parent? But even very particular interventions like whether or not vinegar eaten with a meal will help decrease your glycemic response seems to be one where people are really responders and nonresponder.

Andrew Herr (18:48):

Yeah, I would say that there are a lot of subgroups that we can’t predict based on the demo data, demographic data, and things we’re talking about. So, what are the most likely explanations for the vinegar helps doesn’t help. It’s probably the fact that have vinegar has short chain fatty acid acetate. Some people it probably interacts with some people’s gut microbiome in a way that then prevents or accelerates an inflammatory gut response to something that they’ve eaten. So, you could say, “Well, lets subgroup based on microbiome.” But your problem then is it may be actually the effect of the food, times the effect of the microbiome, times the person’s say susceptibility to inflammatory issues. And so, we think that this is a perfect problem to be made amenable to solving bio machine learning. And so, to do that we’re going to need data sets though that are experimental.

Andrew Herr (19:48):

And so I think that’s where if you can combine running experiments with high fidelity, high frequency data. That to me opens up the future of the field. And so, I think where Levels has made high fidelity, high frequency data available to more people on this critical metabolic parameter on blood sugar, that’s one step to being able to run and better interpret experiments. And then, obviously we’re going to talk later about other things we might want to measure, but as you go from a few things you can measure like that to tens of things, then the covariates and other factors are important. But our big thing is you’ve got to be running experiments otherwise you don’t know what caused what.

Dr. Azure Grant (20:31):

Yes, agreed. 10 points to Gryffindor on all of that. So, let’s start with what measures do you currently take at Fount?

Andrew Herr (20:40):

So, we do a baseline for everybody. So that’s about a hundred blood and urine biomarkers.

Dr. Azure Grant (20:48):


Andrew Herr (20:48):

It’s a 14 to 15 tubes. It turns out that if you work with the major labs in the country, you find out what is the maximum number of tests you can order before it messes their system up. And there is-

Dr. Azure Grant (21:03):

What is that?

Andrew Herr (21:04):

… It’s about 35 with the major players. And after 35 they have a second page in their system that the people don’t know to look at. There’s a really bad UX, like user experience problem, or user interface problem. So, you actually don’t want to order more than that because they’ll actually not draw those labs. So anyways, you end up figuring out how many tests you can order. And so we’re looking at things like-

Dr. Azure Grant (21:27):

Are you working with multiple labs then?

Andrew Herr (21:29):

… We primarily work Quest day to day, but we do in other circumstances work with other labs. But our standard draws is through Quest. And so, we’re looking at inflammatory markers, we’re looking at hormonal markers, we’re looking at metabolic markers. Beyond those liver function, kidney function, cardiovascular risk markers like lipids, and others. So, really going down through all of these different systems, vitamins, minerals, and we’ve really targeted it. I always say we order a lot. I know we ordered a lot of labs, but there’s plenty more we could order, but we’re really interested in the things that are either most predictive or most actionable. But the flip side of that is people always say, “Oh the blood work you order must be the most valuable thing.” And my answer is, “Actually the subjective data is the most valuable piece.” We can run a world class program for somebody without blood work, not nearly as good as we can with it. But you can’t just do it off flood work it turns out.

Andrew Herr (22:35):

But we want to combine everything. I talked about intake interview, so that’s 90 minutes at the beginning for everyone. And then, everyone we’re tracking their sleep data through an Oura ring where we’re getting nightly, not only sleep but also heart rate variability, looking at sympathetic tone as a proxy for stress levels, and recovery, and those factors. And then we’re adding to that now full body MRI, that’s a cancer screening tool. But you also find rather interesting either inflammatory inputs or for people who have various orthopedic challenges, you find things that may or may not explain. Although public service announcement, just because you have a slip disc on an MRI doesn’t mean you have symptoms. In fact, most people have things like that on MRIs that are actually don’t really matter.

Andrew Herr (23:23):

And continuous glucose monitoring, we’ll be rolling out here shortly as well as DNA things. DNA we’ve been really, I’d like to say intentional about because there’s some really tricky privacy and security issues coming out of the military world. We want to make sure we’re really protecting our clients and then also we get the DNA and the epigenic factors integrate with environment to give you the outputs into blood work. So we’re already seeing a lot of that and can back interpret. But obviously we’ll want to be using any data source that allows our customers to get the results that we want to give them.

Dr. Azure Grant (23:59):

That’s amazing. I wasn’t expecting hundreds of things, I was thinking maybe you would have a list to rattle off on a couple hands, but that’s really, really cool. And it also gives me questions about your big list. So, you sent us this big list of your dream biomarkers. And I’m especially interested now since you seem to have found a way to measure those things. Maybe let’s start with the top three favorite things that you have on your big list of dream biomarkers. What are the challenges you found to measuring them that mean you can’t measure them yet? What do they do and what do you want to do with them? And how often would you like to measure them?

Andrew Herr (24:41):

Top three is a tough standard here, but I think the question really was like if you could have continuous monitoring, what would you want? And then there I think about for continuous monitoring, I want to measure things that a point in time measurement is not valuable. So, blood sugar at a point in time is really hard to interpret. You don’t know if you’re going up the slope, down the slope, are you at the peak, are you in a valley. So things that are released in let’s say a pulsed faction fashion, like pulses of it are released and it might blip and then come down. But these matter, so that’s things like stress, hormone levels, cortisol, epinephrine, or adrenaline, that’s hormones. Human growth hormone is released in a very pulse wise fashion. Even things like testosterone and then neurotransmitters. We would love to measure neurotransmitters in at the neuro level in the brain, but I don’t think it’s a good idea to be sampling from the brain.

Andrew Herr (25:36):

I think that’s going to cause more problems right now than help. But I think people under sometimes underappreciated the importance of these peripheral neurotransmitter. So, serotonin is called serotonin of sero-blood-tonin like basically was originally defined as a blood pressure modulator. And so, serotonin, dopamine, all of these things are I think really valuable. And then one I’m absolutely fascinated by, the new research on food sensitivities shows that you can have an allergy to a food only in the lining of your gut and not in your peripheral blood. So that means that most food sensitivities you cannot test for today. The blood tests, the skin tests won’t test for them because the allergy doesn’t exist there.

Andrew Herr (26:25):

And so food sensitivities are such a big issue for people and when we get rid of them by illumination diets for clients, they feel so much better. But there is a systemic effect of this allergic response in the gut where histamine and other cytokines and chemokines aren’t being released, these immune factors. And so, if we could watch you eat different foods and in real time watch the production of things like histamine peripherally, it’s my belief that you could make a continuous food sensitivity measurement that would actually really help people optimize their diet in a very cool way.

Dr. Azure Grant (27:08):

That would be amazing. So you think a combo of something like maybe a CRP and a couple interleukins and inflammatory cytokines would be something that you would like to see be able to rise maybe really quickly, along with a glucose spike after you eat something you’re sensitive to, and then maybe not if you eat something that your body’s good with.

Andrew Herr (27:31):

We’re going to that future, I want that future for everybody, and I think it will be very valuable.

Dr. Azure Grant (27:41):

Yeah, I think that’s great. And you’re mentioning neurotransmitters, especially dopamine is one that I think is super interesting. Difficult in that it doesn’t like to cross the blood brain barrier to my knowledge. But this is one of the things that got me really interested in pulsitile regulation of biomarkers in the first place. I remember reading, it’s many years back at this point, but about dopamine from the vental technical area or VTA directly happening in time pulsitile with the movements that an animal would take. Makes a lot of sense, dopamine from VTA motivating makes you want to get up and get around and do stuff. And then, also learning later that it’s possible that this might be coordinated with this part of the brain called the arcuate pulse generator. Which ends up helping stimulate pulses in a lot of other hormones, particularly in reproductive hormones.

Dr. Azure Grant (28:42):

And, I think that what you mentioned about pulsatility. Meaning A, that if a hormone is pulsitile, it makes more sense to measure it continuously. And then B, that there might be some signal in the shapes of those pulses as something I find super interesting. So, for instance, I’m curious to hear your thoughts more on continuous testosterone measurement, because not only is this something that I think a lot of guys might not be aware when their testosterone starts to decline due to stress or age related factors. But I was also really interested to learn that when there’s a person has a socially successful interaction with another person, for guys testosterone goes up and if you have something you perceive as social defeat, testosterone will actually go down. And obviously I’m not a guy, but I’ve told several of my friends about this and have universally got a response that, “Oh my gosh, it would be very motivating for my husband to know what my testosterone was doing.”

Andrew Herr (29:40):

Yeah. And I think, you just nailed a ton of really interesting points. But one, yes to all of these interesting interaction states and the shape of the curve I think is a nuance point in our field. But a really important one, and I’m sure you’re doing a ton of work on what the shape of the curve looks like for blood glucose and why you would get different shapes. We’ve been doing some internal testing trying to look at that, we don’t have the size of the data sets you have, but this question of why do some people get spikier responses, and why do some people have smoother responses, even if heights and area under the curve, if other parameters are similar, what’s going on there? So, very interested in that question. And then, to testosterone, there’s another level which is if you are high status versus low status, it actually affects whether you get the testosterone boost or you don’t.

Andrew Herr (30:36):

And so, you can just imagine these really fascinating interactions between meditation and breathwork, to mitigate some of the anxiety factors between of status interactions along with CBT and other things. Combined with understanding how different interactions cause you to be more performant or less performant. So I think all these things are interesting and then to your point, the right type of feedback at the right time is highly motivating. And so, I think how do we get the right type of feedback at the right time with the right solutions for the client? So we think, for example, when we look at sleep data, if you don’t sleep well that night, we want to be able to tell you here is the intervention based on how poorly you slept last night, here’s what you should do this morning to perform better during the day.

Andrew Herr (31:27):

And, there’s lots of things we can do around the inflammatory effects of insufficient sleep, of the sympathetic effects of the arousal effects, not sexual, but sort of energy levels and focus things. So, there’s nap timing. Most people are like, “Oh, I don’t have time to take a nap during the day.” Except the studies say maybe five or 10 minutes is optimal. And so again, what are our recommendation sets customized to your body and your goals in your day based on the data I think is motivating and then solve the problems. And then we’re really cooking with the S.

Dr. Azure Grant (32:07):

Totally. That reminded me a little bit of, I mean overall a weird book, but there’s a part in this book called Fall that was one of Neil Stevenson’s recent book where the main character talks about naps and for him the feeling of having broken the thread, whether that took five minutes, or whether that took 20, or 30 minutes, is this thing that kicked him back into action. And so, I’ve wondered how well that had been tested, and if it were really true. I’d also seen you mention some things about nitric oxide, which I think is pretty related to the need for measurement of sex steroids. Is that still one of the top ones on your list?

Andrew Herr (32:47):

Yeah, I think there’s interesting for a number of reasons. One, as you eat things that cause oxidative damage your blood vessels, you tend to destroy nitric oxide. And so, you could watch as you eat sugar that it would probably drop your nitric oxide levels and do other things. And so, we think this damage to vasculature beyond just the cholesterol and cholesterol particle piece is an underappreciated factor for longevity. So we’re thinking a lot about interventions that can protect the blood vessels. This has come even more front and center as we look at COVID and think about it as an inflammatory and potentially a vascular and clotting disease. So I think that, and then also all of the ways that blood vessel dilation is involved in performance and I think that’s performance from the athletic standpoint to performance in bed where you want blood vessels to dilated the right place, the right time to perform well.

Andrew Herr (33:54):

And, one of the funny things you get is if you dysregulate these systems, so for example when people take things like Viagra or these drugs that cause specific types of vasodilation opening of the blood vessels, often people get a stuffy nose while also having better sexual performance. So, that just shows that you want this blood vessels to open in the right place at the right time, can’t breathe while you’re trying to perform, that’s not optimal. But anyways, I digress. But I think the idea of blood vessels as a major performance component is we think underappreciated and also is a longevity component. So I think you’re absolutely right. And then nitric oxide also plays into signaling for all these other pathways that I think you were maybe referring to.

Dr. Azure Grant (34:45):

Yeah, absolutely. I mean it was funny that you mentioned my advisor in grad school actually did some older work on nitric oxide and male reproductive function. And of course it’s this key player for making steroids for sertoli cells, even for helping stimulate ovulation in females. So it sets off these little local inflammatory cascades, and bring down steroidogenesis, and [inaudible 00:35:13] and granulosa cells of the ovary. I mean basically it’s like the super widespread signaling molecule that does so many important things. And, I think it’s really interesting because it’s obviously super useful, but if we were to have a continuous signal of nitric oxide, we would probably have to do a lot of work for ourselves to explain how to interpret it given what was going on with the person you roofed with.

Andrew Herr (35:38):

Also just take liberty of getting one other point there that I think you just really illuminated. I think this continuous monitoring piece is going to be even more valuable for women than men, because there’s so much less research that’s been conducted. Women have a somewhat more complex hormonal cycle. So often basically to exclude variability. They’ve been excluded as research participants, which means we have less data on women. And, if we start to get these continuous measures and you can correlate the effects of different hormones and physiological states and outcomes, I think it’s really exciting you end up actually finally solving some of these problems. And honestly I think it’s the faster, better way than trying to run the same studies on women that we did on men that were only time point in time, let’s go to continuous monitoring of these various things and let’s actually do it and I think we’ll crack the code on some things we haven’t even really thought of yet.

Dr. Azure Grant (36:33):

Absolutely agree. Yeah, learning about that for the first time was one of the things that got me interested in the interactions between reproductive function in women and in metabolism. For instance was learning about type one diabetes in grad school and wanted to understand the impact of the ovulatory cycle on continuous blood glucose. Because progesterone tends to make people a little less insulin sensitive, I mean it affects carb, lipid and protein metabolism, probably direct acts directly on pancreatic eyelets, helps liver store more glycogen, and also stimulates deposition of body fat, and has catabolic effects on protein metabolism. So, it has all these really complex and interesting metabolic effects, but it was really difficult to find data on good well tracked ovulatory cycles, frequent progesterone measurement alongside blood glucose. So, estrogen and progesterone alongside testosterone, are really high on my list of what I think would make a big impact in family health research pretty much across stages of life.

Dr. Azure Grant (37:48):

One thing tying back to what we mentioned earlier about how that part of the brain, the arcuate nucleus helps set up pulsatility in different hormones. One of the really important sets of pulsatility that it sets up via the cascade of gonadotropin releasing hormone GnRH to luteinizing hormone in the pituitary LH all the way down to ovaries, estrogen, and progesterone, is that pulsatility is really important for setting up ovulation. And pulses in estrogen, in LH, and actually also in body temperature, and heart rate variability, speed up leading up to ovulation and then slow down a little bit afterwards. So, that’s one of my favorite examples about how we might be able to use the shape or the speed of pulses in a signal in order to make a prediction.

Dr. Azure Grant (38:40):

But on top of that I think it’s not just about ovulation in the fertility space. I think estrogen and progesterone would also be incredibly important alongside autonomic markers for monitoring pregnancy onset and progression for helping identify early regularities in perimenopause, and maybe even helping as well in cases of hormone addition into the body, whether that’s for a menopause treatment or in people who take hormonal birth control. And I have to ask you now, tell me a little bit more about these PMS learnings or reproductive health related learnings that you guys have had.

Andrew Herr (39:17):

So as you indicated earlier, there are inflammatory signals that are part of these cascades of helping the body communicate. Inflammation is usually used as a catch-all word let’s say, for actions of the immune system that are bad. But we need some of these increases in activity in the immune system to regulate our body. And so, we’ve been looking at, what is the effect at different times in the cycle and how would you modulate nutrition, exercise, supplement protocols, through the cycle in a way that supports the body, mitigates the negative effects on performance, cramps, other factors without messing up the underlying physiological mechanisms and without using really blunt force instruments like hormones or things like that. And so, we have some, I think cool insights into natural products, plant, fruit, and vegetable extracts that have, let’s just say unexpected effects that I can’t talk about yet, but have some unexpected effects and really help in the inflammatory side in a way that is really beneficial.

Andrew Herr (40:36):

So again, what we want to do is we don’t want one size fits all solutions. We want to say, ideally we will hyper customize to use an individual and that’s what we do in our programs and that’s why we have four month long programs, where we run dozens of experiments versus things like that. But if we’re going to productize things we at least want to customize it to your body and temporarily. And so, there we’re looking at using either subjective or we’re working with some potential partners that can do objective tracking of where you are in the cycle. That we will then modulate your protocol based on a few things we know about your levels of vitamins and other hormones and modulate based on the time of the cycle. And one of the things we’ve invested in that’s I think it’s cool to make this all happen is we run our own custom supplement packaging facility. So, we can put together a pack that’s like different for anybody that has this on the morning of day one, but not in morning of day two, it has it in the morning of day six.

Andrew Herr (41:41):

I think we have 80 different items that we can do any level of customization. So, it’s really cool to be able to generate a custom protocol and then be like, I felt great day one, two, and three, day four I felt off, and then change day four only and see if we can really optimize. And then, the program we offer is expensive, because we want it to be the best program in the world. We want to give you a lot of service and experience. Our coaches are former army rangers and navy special operators. It’s a really fun thing to do, but we also understand that not everyone can afford that. And so, we want to make some of this cutting edge insight we’re generating from our clients available to people. And so, that’s why we want to build products out of it. And so, this will be the second product we launch. We have one really cool one on jet lag that we’re launching here, and some people have already been using in the environment. And that’s our goal is to make the really high end stuff available to millions of people.

Dr. Azure Grant (42:41):

That is so cool, man. I was just thinking about personally, I like to do longer distance running and the combination of work going into tracking and ovulatory cycle, trying to look at phase effects on training, learning how to eat differently, and coming to understand this pattern really well. And then doing the same thing talking to older generations in my family or friends who are on birth control. It seems like everybody has really gone into the self detective process, but I don’t know anyone who has really nailed down day of cycle dependent supplementation or really gotten the diet that works perfectly for them. So that sounds amazing and I agree on when you want to push the envelope, it’s going to be expensive at first and if you want to be able to order hundreds of biomarkers or do really specific personally intensive experiments, that’s going to have to come at a cost, at least for the first lot of people who do it.

Dr. Azure Grant (43:44):

But I really like your philosophy of trying to take those learnings and then somehow make them open access. And I think that feels very close to Levels philosophy of trying to respect the data that we collect by making the learnings from it more publicly available and hopefully some of the de-identified data publicly available too. So, we are almost up at the end of our time today. I think the use cases that you mentioned are our super, super interesting. I love that you’re combining blood markers with continuous peripheral data and also interested in the future of continuous blood biomarkers. Is there anything else that you want to bring up that we haven’t had make it on list yet?

Andrew Herr (44:29):

No, I mean I think I really enjoyed the conversation. I would just say this is my opportunity to talk to your audience. And so, my favorite things to harp on are run experiments on yourself. You will find out super interesting things. And in combination of that, if you listen to your body you will notice things and often those are then explicable. I mentioned that if you feel lower energy and more anxiety after a meal, that’s often inflammatory effect, could also be a blood sugar effect. But you could start to find these patterns and then back out what’s causing them. So, love any opportunity to help people experiment. I think things like Levels where you have better data or make it easier to collect data, anything like that where we can make it easier to run these experiments, get data, and help people reach their goals, that’s amazing, and that’s this future I believe we’re all pushing towards. And I think we’re closer to that future than people think.

Dr. Azure Grant (45:32):

Sounds great to me and I love the missing ingredient or the important thing isn’t what you might think of measure everything in the world, its to combine that with a really deep dive into your subjective experience and almost like the permission and guidance in order to pay attention enough and ask yourself the right questions to be able to figure out what works for you. So, thank you so much. It was wonderful to talk to you and we’ll make sure we put in the show notes where people can go to learn more about you and learn more about Fount.

Andrew Herr (46:06):

Such a pleasure Azure. Thanks so much and look forward to continuing to working on these same challenges and hopefully we get some interesting opportunities to collaborate.