GlycanAge Logo GlycanAge is a biological age test paired with expert advice to help guide your wellness.

Understanding Biological Age and its Clinical Applications with Dr Peter Joshi

Podcast published on 3/28/2023 • Show notes written by Vanja Maganjic

“With a single measure of biological age, we would be able to say - it looks like you're doing really well on most of these things, but you need to target this organ system. The hope would be to measure these ages before the disease emerges, and we could prescribe interventions earlier.” – Dr Peter Joshi

42 minutes reading time

Episode summary

The concept of biological age might sound like a new notion, but it’s something researchers have been looking at for 35 years. A person’s biological age can tell us more about them than chronological age because it includes genetics, lifestyle and environmental factors. Since it’s such an individualised metric, it’s important to find a reliable way of measuring it. In the episode, Rina and her guest Dr Peter Joshi describe 11 different biological ageing clocks, the research behind them and why GlycanAge is such a great tool for providing not just an accurate measurement of your biological age, but also actionable tips you can incorporate to promote healthy ageing. Our guest is a Chief Scientist for Humanity Health, a company using digital biomarkers to measure and improve the rate of ageing. 

Subscribe to our podcast
Subscribe to our podcast on your favourite platform! Thank you 🙏

Conversation timestamps

We Discuss:

  • How has our understanding of ageing changed [02:43]
  • What are the hallmarks of ageing [11:15]
  • Explaining the term biological age [14:25]
  • Multi-omics analysis of biological age [20:50]
  • Are biological clocks predictive of chronological age [26:21]
  • The benefit of measuring biological age [43:36]
  • How to improve the biological clocks [46:45]


About the guest

Peter Joshi

Peter Joshi

 Dr. Peter Joshi, a former actuary, and tenured academic at the University of Edinburgh. He is presently the Chief Scientist for Humanity Health, a company focusing on measuring and improving our rate of ageing. He specialises in genomic analysis of complex traits, especially lifespan.

Follow Peter on social media


Humanity Health


Articles, books, and other media discussed in the show

A catalogue of omics biological ageing clocks reveals substantial commonality and associations with disease risk | Aging (2022)

Genomics of 1 million parent lifespans implicates novel pathways and common diseases and distinguishes survival chances | eLife (2019)

Biomarkers of Aging | Experimental Gernotology (1988)

The Hallmarks of Aging | Cell (2013)

Orkney Complex Disease Study - ORCADES: The University of Edinburgh 


Conversation highlights

"In recent years, the biggest thing has been understanding the genetics of longevity, lifespan, and ageing, and to an extent that it's less important than we thought to start with biology than lifestyle. Both choices and lifestyle and the social situation that you're born into, or living in, are much more relevant to how long people live and how well people age." 

"In particular, the thing we were obviously looking for is, could we find a gene that was likely to make you live to be 90? And the short answer to that is no."

"Is there a difference between brain age, heart age, liver age, and kidney age, for an individual? And is your single biological age that we've been trained to measure, essentially the composite of all of these things?" 

"The fact is, at the moment, to my knowledge, biological age tests are not being used in clinics for conventional diagnostics tests, and conventional prognostic tests. I'd say, on the one hand, we need to improve the clocks. But on the other hand, I might argue that when your doctor looks at your BMI, your blood pressure and your cholesterol ratios, and says, your red, green or amber, it's actually a presentational thing that could be presented as a biological age that your heart age appears. You're 40, your heart age appears to be 50, you need to do something about it."


Episode transcript

Rina’s Intro

Rina Bogdanovic [00:04] Hello Hello, and welcome to GlycanHub, the podcast in which we explore health, disease and longevity through the lens of glycobiology. My name is Rina, and I'm your host. Have you ever wondered how come some people who are more advanced in age look younger, while others might appear or feel old for their age? This happens when one's chronological and biological ages do not align, while a chronological age steadily increases with the passage of time, the process of biological ageing is not fixed, and can indeed differ quite substantially among people of the same chronological age. My guest today had a rather unconventional career progression. Starting as an actuary, he now specialises in genomic analysis of complex traits, especially lifespan. Much of his work focuses on exploring how much of our lifespan is actually determined by our genetics, as well as by our circumstances and choices. In our conversation, we address some of the questions integral to our current understanding of biological age. For instance, does each of us have a single biological age, or rather a set of biological ages that might relate to different bodily systems? In which case, what does it mean, if these different biological agents do not align? We'll discuss the various biomarkers of biological age used today, from DNA methylation to telomere length, IgG glycosylation, all the way to digital biomarkers. My guest is a tenured academic at the University of Edinburgh, and the Chief Scientist for Humanity Health, a company focusing on measuring and improving our rate of ageing. A warm welcome to Peter Joshi. 

Peter Joshi [02:00] Hello, Rina. 

Rina Bogdanovic [02:02] So what I found very interesting about you is that you started off as an actuary. And so I'm curious how come you decided to change your career path?

Peter Joshi [02:06] Yeah, I guess. I've worked in financial services for about 20 years. That career has been very good to me, but I thought it might be nice to try something else before I stopped working. And I could just see that whereas, for most of the 20th century, physics and chemistry had been the coming sciences, that biology and particularly genomics seem to be the coming science of the 21st century. And so I decided to find out more about it and get involved and things grew from there.

How has our understanding of ageing changed

Rina Bogdanovic [02:43] Most of your work in recent years has focused on ageing. And as you said, biology has progressed quite a lot in the last few decades. So I'm curious, how has our understanding of ageing changed? 

Peter Joshi [02:57] In recent years, the biggest thing that I've been involved in has been understanding the genetics of longevity, lifespan, ageing, and to an extent disappointingly although, perhaps happily for society, we discovered that it's less important than we thought to start with [biology] than lifestyle. Both choices and lifestyle and the social situation that you're born into, or living in, are much more relevant to how long people live and how well people age. 

Rina Bogdanovic [03:34] Now, I think it's a very common misconception that you know, many people think that if we look at the age at which our parents passed away, that is a pretty good indicator of how likely we are to reach old age. But you're saying that genetics is actually telling a different story?

Peter Joshi [03:51] Yes, and it’s not just that genetics are telling a different story, but you know, working under the presumption that if you're born in the long-lived district of your city or country or vice versa, that that's highly determinative of how long you will live. I mean, there is determination in that I would hasten to add, but at its simplest,-  if your mother has lived 10 years longer than [others] when she dies, you might expect the obvious as well. An assumption is - I could expect to live 10 years longer than everybody else as well. And the short answer to that is no, you could expect to live about 0.8 years longer. But you personally, and obviously, we're talking statistically across, you know, 1000 people or whatever. So longer living parents do have longer living with children, but the longer-living children are only very slightly longer living. And that's certainly telling us that it's not a highly genetic trait, and I think it's also telling us that personal choices and chance are playing as big a role in how we age and how long we live, as much as genetics, or the circumstances we're born into.

Rina Bogdanovic [05:07] Now, if I understand one of your studies really played a pretty big role in our understanding of this, which was the study looking at 1 million lifespans, could you tell me what was unique about this study in particular?

Peter Joshi [05:18] I think the first answer the, biggest answer, to your question is in the question, studying 1 million lifespans, was a breakthrough at the time, a breakthrough facilitated hugely by UK Biobank who recruited half a million British people into their study, asked them questions about their lifestyle, asked them questions about how long they and their parents had lived, and asked them to donate their blood, to extract DNA. And as a result, researchers like me are able to then study the details of which regions of the genome and which genes at its simplest are affecting how long we live. The principal findings of this study are that a little bit as I alluded to earlier, there doesn't appear to be any common variation that is making huge differences in how long people live. And in particular, the thing we were obviously looking for is, could we find a gene that was likely to make you live to be 90? And the short answer to that is no. For two reasons. Firstly, the effect sizes that we discovered of a particular genetic variant being carried by you or me are much smaller than one might have intuitively thought. If you're carrying a favourable genetic variant, in a particular region, and a particular gene, that might lead to your leading 0.1, 0.2 or 0.3 years more life than somebody who carries the unfavourable variant. Of course, there are lots of regions of the genome. So you might be really lucky and you might be carrying favourable variants at many, many, many positions in the genome, you'd be extremely lucky to carry favourable variants at every position in the genome, that's not happening, but you might have more than your fair share. And that may add a year or two to your life in terms of the sort of research that we were looking at, or what we've been able to find in a specific region. The second thing that we found really and here I guess the clinicians might not be overly surprised - the principal thing that seems to affect how long people live in terms of genetic variation is whether or not they're carrying a variant that's making them susceptible to disease. We didn't seem to find any variants that were about generalised ageing, but instead, we were finding variants that were about, say, heart disease, cancer, neurological disease, specifically, and each one of those then contributed to (if you're susceptible to a given disease), how long you might expect to live due to that disease, perhaps shortening your lifespan. 

Rina Bogdanovic [08:23] But the positive thing in that area is that for many of these diseases, lifestyle can actually play a pretty big role in prevention.

Peter Joshi [08:31] Well, exactly. I think, if we think of that fact, it's positive for all of us. It does look and I wouldn't, in any stance minimise the role of chance in all of this. Identical twins usually die of different diseases at quite different ages, you know, their environment, and their genes, are clearly identical. Their lifestyle choices obviously vary, but probably are more similar than yours and mine. And yet they die at different ages, and I would attribute much of that due to chance. But certainly, yes, it does look like the lifestyle choices that we make influence the way we age, how we age and when we die. The most obvious way to see that is the massive increases in lifespan that we've seen since the 1950s. Lifespan, which I’m probably best qualified to talk about in the UK, has increased by 10 to 15 years since the 1950s. A, that's definitely not been about the changes in genetics, whatever people might think, changes in genetics happened over millennia, at its simplest, maybe slightly shorter than that, not 50 years. And so it's not been about genes, it's that we've seen many improvements in lifespan and it's been about the overall social environment. And again, credit to the medics, all of the medical discovery and development that we've seen, most obviously antibiotics, maternal care during childbirth and early infant mortality. But beyond that, we've seen huge improvements in how long people are living. And that's because I think, around 50%, one study we did suggested that people are falling prey to disease at 50% lower rates than they used to, and at the same time, their chances of recovery and the length of time that they would live. About half of the improvement in that - Well, to be honest, it was more recent data, but conjecture, about half of the improvement that we've seen in mortality is people are not falling sick so often, and half of it is because when they do fall sick, they recover better because of medical treatment. And perhaps because they're more robust in the first place, at say age 60, compared to somebody who fell sick in 1970, age 60.

The hallmarks of ageing

Rina Bogdanovic [11:15] Now, our conversation today is going to focus on biological ageing. And there's been a rising interest in measuring one's biological age and measuring one's ageing rate. So I think maybe a good place to start is by defining what the hallmarks of ageing are and what is the purpose of measuring them.

Peter Joshi [11:33] The hallmarks of ageing are really, well there was a paper that Carlos López-Otín authored a few years ago, nearly 10 years ago. And they proposed that they were at a cellular level, what they called hallmarks, six signs of ageing. And they highlighted from my point of view, a few particularly interesting ones, around the genome, in particular genome instability. The stability of the genome, and the way that things are ordered within the genome declines with age. A lot of people have been excited for a long time about telomeres and the length of telomeres, people see telomeres as the protective ends of chromosomes. And what we've seen is that telomere attrition increases with age, essentially, those protective caps are wearing out. We see epigenetic alterations. That's to say your DNA code, well, it stays the same over your life, the cage surrounding your DNA code alters and that's called epigenetics. And we see that altering with age. And then we see other things like cellular senescence, essentially cells going dormant, as you age. And, we see altered intracellular communication, again, as people age. I'm not sure how many I mentioned, they are certainly far from all the hallmarks of ageing. But that's the idea that at a cellular level, we can actually see the relationship between age and cellular function. And a lot of research is really focusing on whether or not those alterations in cellular function are part of natural development. Yes, definitely they are. Whether they're a result of damage to the cells, and therefore over time, we become less functional, less capable - yes they are. To what extent it's an interesting question for each of those hallmarks. And then again, is there something underlying which is ageing? That is directly what those hallmarks are measuring. Or are these merely consequences? If it's directly ageing, if we can intervene and see increased genome stability, perhaps we could actually see the cells then becoming less aged and recovering other functions. If on the other hand, it's just a consequence of ageing, that may not be quite so feasible.

Explaining the term biological age

Rina Bogdanovic [14:25] But what I find maybe perhaps most interesting about this is that these changes as we age occur at different rates in different people. And then this is where the idea of biological age comes in. So could you explain what we mean by biological age and what we mean by chronological age as well?

Peter Joshi [14:41] Sure, absolutely. So this has been a focus of quite a lot of my work in recent years. So biological age is a concept that was first proposed by Baker and Sprott 50 years ago? Not quite. 1988 if I remember rightly. And maybe it's best to start with a kind of real-world example. For any of our listeners who have been to a high school reunion, and I'm sure those who haven't can imagine one. And let's imagine this is for people aged 40, you've last seen many of these people when they were aged 18. Now they're all aged 40 and when you arrive, kind of by definition, or give or take a few months their chronological age is 40. So that's what we mean by age normally, but what Baker and Sprott were talking about was, some of those people will seem biologically, and if you turn up to the high school reunion, most obviously, in terms of physical appearance, different from each party. Some people will have seemed to age very well, and perhaps look not that different from when they were 18, or at least, perhaps, like other 30-year-olds, and vice versa, some of those people might look more like 50-year-olds. Whether that's their skin texture, their hair colour, their posture, or their ear lobe size - you may not be familiar with this, but your lobes grow continuously throughout life. So these attributes that you have biologically, are indicative of ageing. And the interesting question that Baker and Sprott really asked was - if someone has a biological age of 45, but a chronological age of 40 if that's to be truly meaningful, they should look like someone who's 45 biologically. And therefore, if their intrinsic biology is completely of that of a 45-year-old, that biological age should be more prognostic of health and age-related health and age-related disease and mortality than the chronological age of 40. And so we're coming up with something that is closer to the thing that we're kind of really interested in, which is, functionally what age are you rather than what the time on the clock says. So if we could extend that to your attitude to life, as well as your biology, and the two are related, of course, but you know, here, we're interested in the sort of cellular biology, the overall functional fitness of each organ, and how all of the organs are working together, you know, to give you a healthy life, and extending that and biological age is the shorthand measure for doing that. I think the other great thing about biological age is that I think people could relate to it. If I tell someone that your relative risk of heart disease is 1.3, all I get are more questions. Well, is that good or bad? It's bad, it's 1.3 times that of somebody else. So you're more prone to heart disease than someone else. Is it really bad? No, you know, there is a range of these things, and so on. And for example, you know, 1.3 times what is my risk of heart disease, people just don't know. Whereas if I say to someone, your heart age appears to be 45, but you're 40, we can quickly relate to it. Well, yeah, that sounds bad. It's not the end of the world that it's 45 there are plenty of 45-year-olds who don't have heart attacks in the next six months. And so on and so on, but it'd be great to get it back down to 40 or below. And so, it's got this scientific notion on the one hand of functional age, but then I think a very relatable meaning to people who are not used to looking at public health statistics and relative risks and all of the things that we look at in public health.

Rina Bogdanovic [19:21] We have touched upon a bit earlier on the impact of diseases on the rate of ageing, so could you tell us what age-related diseases are and how they impact our ageing?

Peter Joshi [19:32] So at its simplest, age-related diseases are diseases whose incidence increases with age. So probably I'm sure most of your listeners are aware there is a set of developmental diseases that occur very early in life. Those are not the ones that I'm thinking about. It's diseases like cancer, like heart disease, where we see a much bigger risk of that disease to somebody who's in their 60s than somebody who's in their 30s. So that's an age-related disease. Other ones are Alzheimer's disease and dementia, or more generally, it's important to recognise as it says in the question - it's age-related. I'm not saying that age is the only cause, or that if you're in your 30s, you can't suffer from these diseases. But your risk of suffering from these diseases increases substantially with age. And what we're trying to understand is, can we develop measures of biological age, which we find are even more prognostic of these disease outcomes than simple chronological age like we were just talking about?

Multi-omics analysis of biological age

Rina Bogdanovic [20:50] Now, your work and your research, focus on the multi-omics analysis of ageing traits. Could you explain what that means?

Peter Joshi [20:58] Yeah. Sadly, the answer isn’t in the question this time, I’m gonna need to do the hard work. And so we really need to start off with omic, I think. As I was saying earlier, I see that in medical science and biology, progress is just accelerating at a huge pace at the moment. And one of the reasons for that is that we're now able to measure many, many things. And by many things, I don't mean simply what might first come to mind that we can measure height and weight and blood sugar level. What we can measure, for example, is fat in your blood, but instead of just measuring, as people have been for these tests will be familiar - LDL cholesterol, HDL cholesterol, we can measure very, specific fractions of very, specific types of fat in your blood. And so we can measure hundreds of different blood fats separately, and that would be a lipidomic analysis or many lipid analyses, okay, lipid being fat. And so that is omic. So we have lipidomics. There are 1000s and 1000s even 10.000s of proteins in our bodies and we can measure the relative amounts of all of those proteins. That's proteomics. We can measure the immune function on different levels, and so on. The whole omic field in a way has really been kept wide open by genomics, which is where I started, and genomics instead of what would have happened in the 1980s 1990s, where you would measure someone's genetic status at one particular region of the genome. Instead, you could measure somebody's genetic status at millions of regions of the genome. So you've got millions of measures of their DNA. That's the genome and genomics is the measurement of those things. So, sorry, long answer, but that's omics, multi-omics is then saying, well, actually, we can't just measure many, many types of protein or many types of blood lipids, we could do both at the same time. We could measure your genes at the same time as well. And then that would be a multi-omics analysis. So on each dimension, we've got hundreds of 1000s of measurements, and then we've got multiple dimensions, and try to integrate that into one analysis, to try to understand whether those different dimensions are measuring the same thing or something different in aggregate. And to understand if we have different measures coming out of each of these different types of measurement, how we can then best integrate that into sort of one meaningful whole, most obviously, biological age, but it could be something else that one was measuring as well.

Rina Bogdanovic [24:24] And how long has this area of multi-omics existed?

Peter Joshi [24:29] Well, as suggested, I think, really, the whole thing has kicked off in the last 15 years with the ability to make the real transformation is actually a very practical one. It's the cost of doing these things. One could have, in many cases, done these things in the 1980s or 1990s, but it would have just cost so much. But with improvements in our technologies, we're able to measure these things. And I think it's really kicked off in the last 15 years. In terms of single omics, so looking at proteins, for example, much of that has been enabled by genomic technology that then has been translatable into different fields, but not always. And then looking at many, many omics together, that's really very new. It's still expensive to gather this data. It's also certainly for the work that I do, important to gather this data, not just for one person, but for 1000s of people. And so the sort of the study that we used to do the work that we've been alluding to, was called the ORCADES study, a very fantastic study run by a former colleague of mine, Professor Jim Wilson at the University of Edinburgh. And he gathered this sort of information for  1000 of people in many, many dimensions, and we've been able to do, what I would have to confess is probably in a sense pilot work as to where this might go. We probably need the ORCADES study or another study to become even bigger than we have already in terms of the number of participants before we can draw the strength of conclusions we'd like to, but it's a new field, and it's just going to get bigger.

Are biological clocks predictive of chronological age

Rina Bogdanovic [26:21] It's absolutely exciting to hear about it. Another very interesting study of yours, focused on the comparison of different omics biological ageing clocks. So to start off, could you say, what are biological ageing clocks? And what were the clocks you used in the study measuring?

Peter Joshi [26:39] There's quite a lot to unpack there. So, biological age, I hope we've got a sense of already. There are essentially two ways that you might think about going about building a mathematical model at its simplest to estimate biological age. The first would be to look at outcomes. So I think the most obvious outcome is just human mortality and death. So let's build a predictor of death. It sounds a bit like the Grim Reaper, but you know, as an actuary, I guess I was used to doing this many, many years ago, actuaries found that sex, age, lifestyle, smoking, and alcohol consumption, were predictive of death. And they're often actually literally in the actuarial tables, we would have the typical person assessed, and then we would say - “Oh, well, if somebody smokes, their age appears to be seven years above that of somebody who doesn't smoke.”. And therefore, if this person's age is 40, their mortality looks like a 47-year-old. And that feels a bit like what we talked about biological age already. But it's outcome focused. The alternative is to think of measuring things like earlobe size, and simply seeing how it relates to chronological age, cholesterol level, blood pressure, and epigenetics we've alluded to earlier, but the structure of your DNA, the structure of various other things within your blood, how does that relate to age. And so we would call those ageing clocks. And in particular, those are ageing clocks, trained by chronological age. So what we're doing is looking at a large population and seeing essentially something rather useless. Can we predict chronological age based on these measures, and use them to get a little bit more technical, chronological age, to teach us how to make those predictions, and then go to a new population and try to make them again? And the short answer to that is yes. And it's simple. It probably has more forensic interest, taken to extreme than biological interest. Can we from a drop of blood (and no other information) tell what sex you are? Yes, I think that's no surprise to anybody nowadays. We can also tell who your mum and dad are, again, no surprise to anybody if we have their blood as well, of course. But finally, a bit more of a surprise, we can tell what age [you are] to a degree of accuracy. The study was really trying to understand how accurately we can predict biological age from the different types of omics we've just been describing. And then importantly, we went on to try and understand if your blood protein levels are predictive of chronological age in this way. What about your blood lipid levels, and your blood fat levels? Are they predictive in the same way? We didn't study earlobe length, but had we, we’d think it would be predictive of the same length. And that essentially, is what we did in this study. And the short answer is that different things have got different capacities to predict chronological age. Some of them can predict it fairly accurately, indeed (we're talking about plus or minus two years) for somebody in their 40s, or 50s, just based on their blood. As I said, one could imagine that I've not seen an application yet, but one could imagine some application for the forensic context of that sort of technology. But, what we also discovered is that there was a degree of overlap and a degree of difference between the different measures and trying to understand exactly how they all fit together. And then probably most interesting for those of us who are fascinated by biological age, to what extent those age predictions that we're making from your body, are predictive of health over and above the chronological age. And again, we did see some associations there. And that's the nub of the study. The study was probably too small to see anything too categorically about the health prediction, we could say a lot about our capacity to predict chronological age, predict is a funny word in that context. I know. But I think you'll all know what I mean.

Rina Bogdanovic [31:56] If I'm not wrong, you mentioned 11 different biological ageing clocks. And so both when constructing and when measuring, with these clocks, how can you know which ones are measuring something that's actually meaningful compared to those that might be less accurate?

Peter Joshi [32:13] The first meaningfulness is simply - is that particular clock predictive of chronological age, and if it is, it's meaningful in that sense. And so for example, we also looked at just body shape, you can make some fairly sophisticated X-ray-style scans of people's bodies, and see essentially where the different muscles and fat compartments are. They're always in the same place, but what size they are relative to each other? How predictive is that of age, but it’s not very predictive, we found even with quite a lot of measurements. On the other hand, if we look at protein levels or these epigenetics, the structure of your DNA (structure is not quite the right word). But I don't want to get too technical with the audience. But properties of your DNA more than its code, a highly predictive potential of chronological age. So that was the first thing. But then what we found, for example, is that there's an attribute of your cells called glycans that relate to your immune system. We found glycans were not particularly helpful in predicting chronological age. They perform medium amongst a spectrum of 11 that you recognise. But what we did see, perhaps not surprisingly, but very optimistically, was that glycans, where GlycanAge was predictive of health outcomes, not surprisingly, perhaps, because it has to do with your immune system. And we were looking at both of these facets as to suggest, actually, that the more interesting biological ages at least, it seems to me are the biological ages that relate to not just being able to predict chronological age, although that has some technical interest, and in terms of understanding ageing processes, but the ones that are then actually more predictive of health as well. 

Rina Bogdanovic [34:39] So we've said you had 11 different biological ageing clocks, each of which was used to measure biological age. What happens when you put all of that data together into one big ageing clock?

Peter Joshi [34:49] That's a good question. So that was one of the things that we did. We said, well, as I was saying earlier in the conversation, you know, effectively, we measured 11 different dimensions of, these omic attributes of your body, and hundreds of measures within those. What we then did was we went on to think about, well, what happens if we put all of that into what we call a mega omic analysis. And perhaps not surprisingly, given that each of them was quite predictive of chronological age, in the sense that we've been discussing when we put them all together, we get something extremely predictive of chronological age. And so, again, in the forensic context I was talking about - if for some reason, someone didn't know or refused to tell us their age, we might be able to infer their age, using these sorts of techniques. Inevitably, if we can predict someone's chronological age, exactly and do that, it doesn't tell us much about their biological age, as distinct from the chronological age, because all we've got is a chronological age recapitulated. So we found it's possible. But it's helpfulness to a clinician in the clinic, where the person is probably telling them clearly what their age is, and if they're sceptical, they can look up their health record or their birth certificate. It's on the face of it going to be less useful.

Rina Bogdanovic [36:23] Now, when thinking conceptually about biological age, you pose a very interesting question in the paper, and I have it here. So it said the notion of biological age raises fundamental questions. Is there one biological age for a person or a set of biological ages, perhaps relating to different bodily systems? And so I'm wondering if, indeed, there are different biological ages, which relate to different bodily systems? Might we overlook potential health risks which exist outside the confines of the body system, to which a particular biological age clock is receptive to?

Peter Joshi [37:05] Yeah, exactly. So I think research to be honest, is limited in this context. And this is why we raise it as a question. But the notion I think is intuitive and certainly does merit further research. Is there a difference between brain age, heart age, liver age, and kidney age, for an individual? And is your single biological age that we've been trained to measure, essentially the composite of all of these things? A composite, perhaps weighted by the relative importance, at the end of the day, of course, all organs are probably, or all those organs, at least crucial to life. And therefore, a failure of one is a big deal, even if the others are healthy. But that's the notion that we're trying to raise. I'm a great believer in it, as a way of trying to disentangle this rather more complicated, or composite biological age that we've been discussing so far. And I think also disentangling any one of those ages between the effect of your current chronological age, the effect of your genes, the effect of your social environment is possibly somewhat beyond your or to a great extent beyond your control. And then your individual life and choices within that social environment, and therefore seeing it's going to become quite a big table and to show it to patients and customers. We need to make it simpler than that. But I think that sort of notion underpinning, a single measure of biological age, and therefore at its simplest, being able to say, well, actually, it looks like you're doing really well on most of these things. But you need to target this organ system. Perhaps most obviously, if it was your liver, drinking less, but you know, that there's, other organ systems, while there are a lot of other organ systems and other interventions in that context, that could be focused, and in a way that's not that different from what medics do at the moment. But the hope would be perhaps if we can measure these ages before the disease emerges, that we can prescribe interventions earlier. Hopefully, interventions without side effects, like exercising more, drinking less, and so on and so on, really motivate people to do those and do it in a targeted way that reflects their susceptibilities and their lifestyles.

Rina Bogdanovic [39:46] Yeah, absolutely. I think the prognostic potential of biological ages certainly sounds very inviting. You are currently working as the Chief Scientific Officer at Humanity Health, which is a company that has developed an app, allowing users to measure their rate of biological age through digital biomarkers. Could you explain what digital biomarkers are? And how do they compare to the ones which we just talked about?

Peter Joshi [40:12] Absolutely delighted to. So yes, that's exactly what we're doing at Humanity. We're measuring people's biological age, we're measuring the rate of ageing, which we'll maybe get to in a moment. And I think as crucially, we're then talking to people about how to adapt their lifestyle to improve their biological age. And digital biomarkers, well, at its simplest, most of us nowadays, have a mobile phone, and we can use the mobile phone and the digital watch to measure biological age in the way that I've been talking about through omics. In that context, particularly, we do other measures of biological age, like blood profile biological age, and epigenetic age at humanity as well. But we're very much focused on these digital biomarkers, essentially arising from movement patterns, and heart rate patterns in the context of those movement patterns to understand someone's biological age or to estimate someone's biological age. And the surprising thing is that one of our partners showed that the biological age measures that we're coming up with, through digital biomarkers is as accurate as that could be maintained from blood profiles so that listeners who have had their blood taken and had their doctor come up with a generalised blood profile of HDL cholesterol, LDL cholesterol, and the various other things that come up in those profiles, we find that using the UK Biobank study that we talked about earlier, there were people who did these digital biomarkers, as well as the conventional blood biomarkers, that the digital biomarkers, were giving us just as good biological ages as the blood biomarkers. And so a few benefits to that - the first and most obvious one is back to my highlighting the strength of this study that we did into genetics with a million lifespans. Digital biomarkers are easier to gather, both as a customer and as a researcher. And so we've got a lot more volume of data. But also, those digital biomarkers are potentially available, more realistically from our point of view daily. But hourly in principle, minute by minute, arguably, but for our use point of view, daily measures that we then make into a composite measure over weeks. And it's easy just to gather them continuously at volume. And then if we see changes in those markers, they become apparent without, for example, having to visit your physician. But also for me, as a researcher, I'm able to see the particular effects that particular customers have and changes and actions on changes in those biological ages.

The benefit of measuring biological age

Rina Bogdanovic [43:36] And what is the benefit of being able to measure the rate of biological ageing?

Peter Joshi [43:43] If we can capture biological age, it sort of works at two levels. One, we're going to be able to advise people individually of their biological age. Secondly, what we see going forward is that we're going to be able to say to someone with certain characteristics of age, weight, sex, and so on - your biological age seems to be better or worse than it should be based on your date of birth. Secondly, what we've learned from people like you is that this particular intervention seems particularly effective. So please try to do an extra 3000 steps going forward or please try to have a longer period between meals than what you're having at the moment or snack less. And that recommendation will have two strengths. One is that it will be effective for the person. But secondly, we're hoping and expecting that we'll be able to come up with recommendations that are practical for the individual. And again, practical not just for people generally but practical for people like them.

Rina Bogdanovic [45:05] Yeah, I think one of the main strengths is being able to develop a programme or a system which gives tailored advice to each individual. So, as I understand you've had a pretty mathematical approach in designing that programme.

Peter Joshi [45:19] Yes, that's right. I mean, it's the recommendations that our system comes up with, at the moment, that was based on a deep review of the scientific literature as to what seems to help and not help. I'm sure listeners regularly read in the newspapers about the beneficial effects of or not of caffeine, of bacon, or exercise, and so on and so on. So what we did was take lots of scientific literature and look at the evidence in conjunction, sleep is another one. For example, you know, we see the beneficial effects of sleep, and sleeping the right amount. But going forward, as I say, because we see the biological age for our customers, and we see the actions that they've taken, we're joining the effect of those actions, and in recent months to the effect of the change in biological age, and concluding, actually, you know, where the scientific literature is robust being born out in our data or not, and therefore adjusting for our customers, both as a whole and individually, the sort of recommendations that we make, as to how to reduce their biological age.

How to improve the biological clocks 

Rina Bogdanovic [46:45] Now, we've talked just previously about the prognostic potentials of using biological ageing clocks in a clinical setting. And I think what really has to be taken into consideration is that these clocks really need to become more practical and more easily accessible. So what do you think are some milestones we have to pass before these clocks can actually become clinically useful?

Peter Joshi [47:09] Yeah, I mean, it's a very good question. The fact is, at the moment, to my knowledge, they're not being used in the clinic - conventional diagnostics, tests, and conventional prognostic tests are being used in the clinic. I'd say, on the one hand, we need to improve the clocks. But on the other hand, I think, I might argue that when your doctor looks at your BMI, your blood pressure and your cholesterol ratios, and says, you know, you are red, green or amber, it's actually a presentational thing that could be in the way that we discussed earlier, presented as a biological age that your heart age appears. You're 40, your heart age appears to be 50, you need to do something about it. And anybody who's more than, you know, 3 years, different from their chronological age, that's something that the doctor remarks upon. And if it's adverse, says “we need to do something about it”. And that in itself is quite translatable. I'm very conscious that the medical profession is quite conservative, and they probably feel they have a mechanism at the moment that works quite well. And we'd be reluctant to change it. I'd question that slightly when you see these tables when you're sitting with your doctor, and it's got percentages on it, as it matches up to have faced by that, but I think many people are kind of left with doubts as to what that percentage means, and therefore, unsure what to do about it. And in a way, what I'm seeing is that coming up with a readily usable measure of biological age that's communicated well would be one way to take forward clinical practice. And then the other obvious way for someone like me who spends their time researching how to make better biological clocks is to come up with something that is more accurate than the existing tests. And therefore, once proven, once communicated well to the community be the natural thing that people will want to use, as opposed to some of the existing tests that are out there.

Rina Bogdanovic [49:42] I think also one question I'm thinking of is that with all of these tests, you've mentioned a lot of them take into consideration a very wide number of biomarkers. So when thinking about making them clinically useful, do you think their reliability and accuracy decrease with a decrease in the number of biomarkers?

Peter Joshi [50:02] That's a really good question. And a second part of the question had a premise that actually wasn't born out of our research. What my student Erin MacDonald Dunlop found was that she could identify a substantial (maybe that's a much smaller set) of biomarkers than the 1000s that we've been using and that was equally prognostic of biological age, as the larger side, and therefore, it does look to us like once we can get large enough datasets that we can make good predictions, and the problem essentially is, you know, if we want to predict heart disease, sadly, we need our study participants or many of our study participants to have had heart disease, that takes time and it takes numbers. So over time, we will be able to gather this information. And then what I'm seeing is at that point, having built a good predictor of maybe using 1000 predictors, we would expect to be able to reduce that to perhaps a dozen. And then, as you alluded to the cost, the feasibility, the speed at which a lab could do that test, and return it to the doctor becomes practical in terms of clinical timescales. 

Rina Bogdanovic [51:37] That's certainly very encouraging to hear. So you have mentioned that our expected lifespan has increased over the last 50 years. But when it comes to these rare communities, such as Ikaria in Greece or Okinawa in Japan, which consistently continue to have longer lifespans than what we observe in general, is there something specific about their lifestyle or their surroundings that allows them to enter into such old age? And probably more importantly, is it something that we can learn from and replicate in our life?

Peter Joshi [52:19] I don't know where to start there. I mean, the short answer is, I feel, I know quite a lot of people think they do know, but my suspicion is that there are just so many differences between those communities and other communities, that pinpointing which of them is the important difference is very difficult. I am reasonably convinced that it's not genetics. Some of these communities have been quite isolated in the past and therefore may have significant genetic differences. But my gut feeling is still that that's not a major part of the explanation. And the problem is, you know if we're to take the difference between, say, France and the UK, and the difference in expected lifespan between France and the UK, which isn't that much to be fair, which of the multitude of differences between France and the UK are we going to take and even if we studied everybody in France together, and then compare that with everybody in the UK together? I think we would struggle. And therefore what we've tended to do is look within populations, like the UK and UK Biobank, and try to, essentially, it's a bit more sophisticated than this. Determine people with a single difference between them, and which one of those is living longer, and then do that 1000s of times a day, and use that as the way to find out. That's not to deny the other part of the premise of your question, which is that there is some secret sauce secret, sadly to people that are there as well as the rest of us as to why these people are living longer. But it's just very difficult to see why those people are living longer than people say in another part of Italy or Japan because there are so many differences. And hence it's the things you'd expect: diet, lifestyle, social support, the socio-economic situation more generally.

Rina Bogdanovic [54:35] Now we talked about age-related diseases and the use of biological age, to potentially improve our health and optimally decrease the rate at which we age and potentially also increase our lifespans. So what would you think would be some of the most obvious societal and economic impacts of slowing the rate of ageing?

Peter Joshi [54:59] There's most obviously the cosmetics  industry, you know, that is targeting the slowing down of the appearance of ageing at the very least. It's something that people want so I think it's going to be a product or set of behaviours with a huge demand from society. I think your question has tried to go a bit deeper than that. I think the intrinsic benefit is, you know, Baker and Sprott’s biological age was all about functionality, we see so much function, not all function, decline with age, cognition does. I believe I would say this does not decline until really quite late in life, a bit more generally, people able to be more economically productive on the one hand, but more importantly, have better health and do things that they want to do, that age might otherwise have been standing in the way of. And if we can, we can slow down ageing. If we can slow down particular attributes of biological ageing, we'll have all of those benefits, and hopefully, less need for treatments, less need for medications, and so on and so on. And if done well, I'd very much like to think of people not just living longer lives, but happier lives.

Rina Bogdanovic [56:27] And lastly, my question really has to do with your view of what you think should be done to extend this field in the future. What do you think are the biggest directions that research should take in this area?

Peter Joshi [56:42] It's sort of research that I've been very involved in until recently, as I say, getting those 1000s people in Orkney, they're already trying to expand that study. And there are more than 1000 people in the study now, but not always the money to measure the many things that we've measured for the 1000 people. We're measuring quite a lot for the others, but you know, at its simplest, more funding to expand those studies and come to firm conclusions and drill down to levels of details within those conclusions. I'm a strong advocate of the digital biomarkers that we were talking about, I think, more surveying of digital biomarkers and health outcomes than the way the UK Biobank has done over time that's naturally coming with UK Biobank, for example. So it's going to get better and better. And then I think, to the extent genetics has succeeded in this, but it's not succeeded as much as we hoped, trying to understand the biological mechanisms that are going on. So where we're measuring biologies, whether it's blood proteins, or whatever, and we're seeing that affect ageing and disease outcomes, trying to understand the biological pathway down which that act, and then build interventions along that pathway as a classic method of drug discovery and drug treatment. But you know, how are the lifestyle interventions, you've talked about these exceptionally long-live communities around the world? What is going on there? What could we replicate in our society that would give us those benefits? I think, in the end, we will be about as much understanding the biology as simply saying, well, it's that particular intervention.

Rina Bogdanovic [58:38] Excellent. Thank you very much, Peter.

Peter Joshi [58:40] Thank you. Rina, it's great to talk to you.

Rina Bogdanovic [58:43] Now speaking to our listeners, I hope you gained a better insight into our current understanding of biological age, as well as its practical applications in preventative medicine. If you would like to access more information about this conversation, Peter's research and Humanity Health, follow the link in the description to the show notes for this episode. Equally, if you would like to find out more about GlycanAge, head on to where you can access a whole list of our scientific publications, blog posts, testimonials, and of course, this is where you can order your GlycanAge test kit. Watch out for our next episode, where I will be joined by Hudson Freeze, a glycobiologist and biochemist whose research focuses on congenital disorders of glycosylation, where we'll be discussing the role of glycans in normal physiological function and the consequences of glycosylation defects. Please don't forget to leave ratings and reviews for this episode, and engage with us on social media. Thank you for listening and have a great day.


Subscribe to our podcast
Subscribe to our podcast on your favourite platform! Thank you 🙏

Start or continue your GlycanAge journey

Don’t be afraid to reach out to us and ask questions, provide commentary or suggest topics.

Other episodes you may like:

Blog image Season 1

The Link between IgG Glycans, Aging, and Disease: A Discussion with Dr. Olga Zaytseva on Glycosylation Regulation

Read full article
Blog image Season 1

Glycans and the Immune System: Understanding Immune Regulation with Prof. Salomé Pinho

Read full article
Blog image Season 1

The Past, Present and Future of Congenital Disorders of Glycosylation with Prof. Hudson Freeze

Read full article
Blog image Season 1

Glycans and Cardiovascular Disease: Understanding Modifiable Risk Factors and the Power of Lifestyle Choices with Prof. Samia Mora

Read full article


Please be advised that this show is for information only and should not be considered as a replacement or equal to medical advice.