Jo Holbrook

5 Letter sequencing & tangible improvements to health care

Cambridge Epigenetix’s former Chief Scientific Officer, Joanna Holbrook, is interviewed by the Precision Medicine Forum podcast.


Voiceover 0:00
Welcome to precision medicine Forum Podcast, chatting with patients healthcare, industry and research professionals about creating personalised medicines for each and every one of us. Together we had to the Holy Grail, mainstream precision medicine. Here’s your host, Steve Caldicott

Steve Caldicott 0:19
Hi and welcome to Precision Medicine Forum Podcast. I’m joined today by Jo Holbrook, the Chief Scientific Officer at Cambridge Epigenetix. How you doing?

Jo Holbrook 0:29
Hi Steve, I’m very good today. Thank you.

Steve Caldicott 0:31
Can we do for genetics, epigenetics, epigenomics? Can you give us a sort of an Idiot’s Guide, if you will, a very brief introduction as to how you would define epigenetics to the layperson.

Jo Holbrook 0:46
So epigenetics basically means all the control of DNA, which isn’t the four genetic letters, ACGT, but all of the chemical modifications that occur on top of DNA and RNA. So these can be chemical modifications to the DNA and RNA themselves, or to the proteins that wrap around them. But what it does is it controls the way that nucleic acid works. So every cell in your body has the same DNA sequence. But they behave differently. You have skin cells and liver cells and heart cells. That’s epigenetics. So it defines cell fate, what type of cell you are, but also how aged that cell is from just your epigenome. You can tell someone’s age, within two years. And it also allows your our bodies to respond to the environment. So you will see epigenetic changes that occur because of our, the way we live our lifestyle, the toxins that were exposed to. And that’s especially important when you’re thinking about response to drugs, because you can see response to a drug to treatment in your epigenome, and that has lots of applications in precision medicine. So that’s where the genome is, it’s especially we know, it’s a lot about changes to the epigenome and cancer. Because as I said, the epigenome is about cell fate. And cancer has changed cell fate. So the epigenome changes a lot in cancer also changes a lot when cancer is treated. And you can look in the epithelium to see successful or unsuccessful cancer treatments as well.

Steve Caldicott 2:29
Whenever we talk about precision medicine. There’s two things that I think two words, the standout for many people, one is genomics, the poster child, everything’s about genomics, although more and more people are talking about other omics technologies. The other thing is, I would say cancer and cancer very much being the poster child of precision medicine. So from where we are now, or where we’ve come from, where we are now and where we’re going? How will it be genomics Do you think fit into that sort of landscape? will it just be? Will the utilisation just be within oncology? Is, is the first sort of question and how does it integrate with genomics per se?

Jo Holbrook 3:18
I’m gonna answer those the opposite way round. It’s part of the puzzle. So what we know is that the epigenome interacts with the genome with genomics. And vice versa. You can’t really understand biology unless you look at the multiple layers of control of nucleic acids, and that means genetic information, and epigenetic, and all those different layers and layers of epigenetic information as well. Okay, rejection ethics, we think a lot about DNA methylation. So the chemical changes that occur to DNA sequence themselves. And you can see that they are they interact with genetics. So I said that DNA methylation changes with your environment that’s dependent on your own epic on your own genetics. So you and I might have different genetic code, and we will respond differently to the environment. And so we will have different epigenetics as well, because of that. So we’re learning I think projects like the ENCODE project taught us this, that we need to look at multiple layers. So all of that all those things that genomics didn’t explain, will be eventually explained by the combination of genetics and epigenetics. And we are at the cusp of sort of understanding all those different layers, and now we need to integrate them together. So genomics isn’t going away, but it’s going to be hugely enhanced by knowing the epigenetic layers. And then to your second question, we know most about the application of Oncology at the moment. That’s true, I think, for the position medicine field in general, and probably genetics in general because cancer is a genetic disease. Actually, you get to the Matic DNA Changes in cancer, that’s been the pioneer for precision medicine. And similarly, we know more about epigenomic changes in cancer than we do in other diseases. So there will be huge applications there. But I also think that we will see massive applications of epi genomics, particularly in all other disease areas, because again, disease areas that maybe aren’t driven so much by somatic mutation, we will see much bigger effect sizes for the epigenome disease, with a huge amount of unmet medical need is neurodegenerative disorders. And there we see DNA methylation and hydroxymethylation changes, which specify neuronal cell fate. And I think that’s, that will be a growing area in the next few years.

Steve Caldicott 5:48
If I’m understanding you correctly, we can very much monitor lifestyle changes. Yeah. Epigenetics, epigenetics, does it there for hold some promise, perhaps in terms of healthcare monitoring? And you know, people go for M OTs, and in full body scans, and what have you those that can those that have the luxury to be able to afford it, of course, could that potentially play a role in that sort of environment?

Jo Holbrook 6:14
Yes, I think so. It’s, there’s certainly epigenetic marks that associate with later risk of disease. I think some of the most famous are the Horvath clock. So Steve Horvath defined a set of DNA methylation marks that associate with our chronological age. So if you look at your blood, you can tell your age within two years from your hofferth clock methylation marks, and then how accelerated or decelerated your ages compared to your chronological age. So if you think of the DNA methylation marks as your biological age, if you are much older biologically than your chronological age, that’s a bad sign for future health. And the Horovod clock has shown that your that that difference is related to all cause mortality and morbidity. So it’s predictive? I think the big controversy in the field is, is it causative? Or is it an effect? So if it’s just effect, it’s still a very good biomarker? But would intervening with switching those epigenetic marks back earlier in time? Would that be would that have a phenotypic effect? There’s some very good work, some of it from bullfights groups, just seeing that there is a functional effect. But that’s, I think that’s where the field is heading.

Steve Caldicott 7:50
Where are we now in terms of how it’s you how it’s used, where it’s used, where the technologies are, in terms of, you know, is it very much as if currently very much a research tool? Or is it being used within the healthcare setting at all? Geographically, you know, obviously, we’re in a position where precision medicine initiatives and genomic you know, sort of large scale genomics programmes and initiatives are becoming more prevalent, but perhaps in countries where they’re, you know, the healthcare systems are at a certain stage. So where are we with, with epigenomics,

Jo Holbrook 8:33
there’s multiple applications, and it’s being layered on to those genome sequencing projects that you talked about. One area that I’m really excited about is liquid biopsy. So being able to read the DNA, the cell free DNA and blood as a marker of disease. And there’s a lot of very exciting research and clinical work now, seeing if we can detect cancer in blood. So if you can diagnose cancer early, just from taking a blood sample, which is obviously a lot less invasive than most of our screening technologies now. There’s multiple, well funded biotechs, who are progressing quickly to clinic with detection of early cancer, and a detection of recurring cancer from blood. A lot of those rely purely on epigenetic marks to do that to the epigenome of the selfie DNA, because, again, it’s dynamic, it’s changed by environment. It also tells you just from that blood sample, where the tumour is, because the epigenome, of course specifies cell type. So you can look at the epigenome of selfie DNA and blood and say, Do you have cancer? And if so, where is it? That’s really exciting. It’s it’s not there yet. And we need some technological advances as well. That’s something that came with genetics works on being able to measure those important genetic and epigenetic Mark’s in tiny inputs, because you don’t get a lot of cell free DNA from a standard blood draw. But that’s a very exciting application of epigenetics. And there were many more in the fields of either diagnosis or treatment response. If you could take from blood or available tissue, add in a sample and epigenomic sample that told you is a drug working for you personally, before you necessarily have symptomatic or phenotypic effects, you could switch patients to drugs that do work for them, or keep them on successful treatment. And that’s really important for cancer patients who, you know, shouldn’t be exposed to side effects for drugs that aren’t working for them, and need speedy decision making on their treatment plan.

Steve Caldicott 10:52
So you say, you said about Cambridge epigenetics and the technologies? So what are the what are the other barriers at the moment, would you say in terms of in terms of a sort of greater adoption,

Jo Holbrook 11:04
apart from technological barriers, so there’s, that’s just the measurement of these marks. And then there’s the integration of those different Omega levels. So at the moment, we have tools that measure each level, we need to be able to integrate those together, either in the lab or in silico in computational methods, that’s technically challenging. And then there’s kind of regulatory hurdles. epigenomic data is different to the genetic data that has been, as you said, at the forefront of precision medicine, it’s qualitative. It’s a sliding scale, rather than a sort of binary, the variant is there or not the genetics. So that means that we need different frameworks, for regulators, things like early screening, for cancer, that is a huge new decision to make in terms of the cost of early screening, it could save health systems a huge amount of money. But you also need to be very, very specific, because you can’t tell people they might have cancer. And it turns out, they don’t, right. So you have to you, there was a lot to be worked out in what is effective clinical pathways in different tumour types, some tumour types, there’s a quick and non invasive secondary screen to check, and some there isn’t. So there’s a lot of work, I think, on the regulatory and clinical pathway for these type of new capabilities.

Steve Caldicott 12:40
It’s interesting what you said over screening, we had a chat or had a podcast a little while ago with Dr. Laura Esserman. She’s out of UCSF, working on the wisdom project, which is really looking at more precise screening, particularly specifically in breast cancer. So for example, the suggestion that why would you be screening a perfectly healthy, no familial history of breast cancer 40 year old woman. And at the same intervals as someone who does have family history, which seems kind of ludicrous, a waste of money, not nice for everybody. And that kind of fits into just what you’re saying, I guess if we can get to that stage where the screening is more precise,

Jo Holbrook 13:34
and less invasive, I think there’s a difference between a blood test and a mammogram. And so you might put up with different levels of sensitivity and specificity for different levels of invasiveness, you know, similar for colonoscopy or a prostate exam, or, you know, I think, very dependent on what the screening is. There’s in terms of both how invasive it and unpleasant it is for the patient, and also how expensive it is for the healthcare system. different sensitivities and specificities are required, and blood testing and non invasive, but there’s still for different cancer types, I think there will be different goals for the accuracy of the test. And again, I don’t think we’re quite there with accuracy yet. And we need the kind of technological advances that Cambridge action ETICS is working on to get there.

Steve Caldicott 14:30
I’d like to get everyone’s view on this because I think it’s important, although some the answers are wildly varied. When you talk about how we can be, we can make it more cost effective, we can make it cheaper and so on and so forth. I think the Holy Grail is that is to find the person who can explain to governments and healthcare systems that precision medicine and all that surrounds it is better, more Effective, better outcomes and cheaper? How do we encapsulate that and present that healthcare systems, to insurers to governments to say, actually, this works and is better and is cheaper?

Jo Holbrook 15:21
Do you feel that that’s messages not accepted a governmental level?

Steve Caldicott 15:26
I think what I see is and consider the factors of city. I’m neither a scientist nor healthcare practitioner, but I do talk to a lot of them. There’s these different approaches with precision medicine, whether it’s a top down approach, or the bottom up. So for example, Sweden, whether they’ve pushed it from research up through into the healthcare system, and then and then it’s coming to government level. One could argue that, for example, the NHS and genomics England is a bottom down approach. You know, when David Cameron announced the 100,000 Genomes Project back in 2012, whatever it was, so who’s, who pushes it? You know, who should push it? Should it be researchers and healthcare? Should it be government? Either with the different healthcare systems? Obviously, there’s a massive challenge. You know, if you look at the, those examples, for example, like the US, where people have got insurance or haven’t got insurance, and one might argue, for example, that precision medicine makes it even less equal. But I’ve spoken to health economics, sorry, health economists who seem to have the answers. But there’s no definitive This is it, we waste a lot of drugs on people, we, you know, we we make people go through therapists that perhaps have no effect. And yet the technologies are there now. We can crunch the data,

Jo Holbrook 16:55
I feel your frustration, because I’m frustrated. Yeah. I felt that frustration most of my career, I started in drug discovery, and development. And I worked on targeted therapies in oncology where we knew preclinically the tumours these drugs would work on tumours they weren’t. And at the time, this is 1015 years ago, it was a struggle to get that even into the clinical trials, nevermind into clinical practice, it has moved on since then. I’m not sure it’s a case of all these different bodies accepting precision medicine as a concept. I think it’s individual empirical cases. And again, scientists, driven by data, pharma companies are very bought into precision medicine now, because their drugs will fail clinical trials, if they don’t pick out responders. And so they are motivated to make companion diagnostics and go to regulators with companion diagnostic alongside their drug because they’ll get better efficacy data. Similarly, health and the payers also don’t want to pay for drugs that aren’t working for the majority of people. And so they, there is acceptance for Pacific drugs to desegregate the population, the patient population into those that are going to respond. So I think there is there is a coming acceptance, but it’s it’s drugged by drug. And where the lag factor is, I think, is that a lot of our processes need to change the regulatory processes, and also the clinical practices, the ability for us to measure these biomarkers in clinic. Often it’s different techniques than we use in discovery. We need to have very effective very cheap diagnostics. That it that infrastructure is also being built. But I think it has lagged behind, you know what we know in the lab, to what we can do in clinic, and that takes time to come through. A lot of these biomarker panels are complex, and they take machine learning to make a prediction, the regulatory path for a machine learning algorithm is still being worked out as well. So I think I can see it coming and I can see it coming piecemeal. In in particular, empirical examples, I don’t think, you know, one of the major health insurers is going to turn around and say, we now want everything to be precision medicine. I think they’re going to approve and paid for specific drugs, which come with a companion diagnostic, or they’re going to start screening tests for Apalis populations, because they can see the hard headed commercial benefit for collaborative treatment. And I think we have to make those cases on the data and the data that was coming through

Steve Caldicott 19:59
there That’s one of the sort of the best answers I’ve ever had to that question. Used to be how the economist Can you can explain to some governments that that’s why we should do it. I really like science. Brilliant. Let’s talk about Cambridge epigenetics specifically, I, you know, I don’t like people give them big sales pitches on our podcast, but we need to know you know, who you are, what you’re doing, et cetera, et cetera. Now, when I look at Cambridge, epigenetics, I see a company that was founded I think, in 2012. And all of a sudden, recent relatively recently, you get huge funding. And it kind of begs the question and Frick and don’t answer this if you can’t, or don’t want to miss the question. What’s been happening for the first nine years? Okay, I can I can answer that question. Why suddenly

Jo Holbrook 20:58
to Cambridge, if Netflix as you say, was founded 10 years ago, the co founders are Shankar bellissimo Omnium and Bobby Well, so huge pedigree in sequencing I mean, Shankar and his and his collaborators invented sequencing by synthesis and NGS. Bobby’s information personnel Chief Strategy Officer of Microsoft. And they founded Cambridge actionetics to build tools and technologies to measure the epigenome. And the market in particular, they focused on at the beginning was five hydroxy, methyl cytosine, okay says huge amount of expertise there, they’ve been producing products in that space. But at that time, epigenetics was a relatively niche interest. All of the stuff we’ve been talking about just now about the huge promise of epigenomics in precision medicine, that’s been emerging over the last 10 years. During that time came shepherdess, Netflix has been building this expertise in molecular biology and the entomology and knowledge around these epigenetic marks, they are showing that some of them are very powerful to detect cancer and early, early cancer in blood samples. For instance, there’s published work by camperships, genetics. And as the emergence of the new applications of epigenomics. And again, I would say liquid biopsy is a very busy one at the moment and liquid biopsy is going to work, we’re going to be able to detect disease in blood samples. Huge amount of research dollars going into at the moment, came genetics, we realised we were sitting on the technology, that’s going to enable all of this because we have a platform that allows you to measure multiple omics layers, so genetics and epigenetics together, and you can do it on existing platforms, you can do it using the sequences you already have, the computational setups you already have. And that’s how we raised the money. Because investors saw the huge potential to empower all of those applications of precision medicine, that are now becoming Prime Time with a technology that’s matching the needs of those applications. So we were busy, for the full 10 years. We built a war chest, which is now exactly what the Precision Medicine world needs. And that’s how the investors were convinced,

Steve Caldicott 23:37
exciting times for sure. It is we’re about to commercially launch

Jo Holbrook 23:41
our first product from this new platform. And that’s exciting.

Steve Caldicott 23:45
My next question. So where are you? Where are you at? When? Where are you at once? When are you going to be there? I mean, obviously, if everything’s okay, but when do you go to when do you go to the world and say this is it. We’ve got the first technology available

Jo Holbrook 24:02
early next year. So we have a technology we call five letter sequencing that measures ACGT methyl cytosine and hydroxy methyl cytosine in the same workflow, a same sample gives you a huge information content benefit from your DNA. That’s with very Aug researchers, very exciting labs across the world at the moment in Early Access, and it’ll be commercially launched at the beginning of next year. And then we have other products, for instance, six letter sequencing, seven letter sequencing coming down the pipe and will be launched pretty soon afterwards.

Steve Caldicott 24:42
Okay, so that’s next year, where where are we going to be in 10 years

Jo Holbrook 24:46
as a field. Precision Medicine will have come of age. Absolutely. We will be everything will be multi omic. We will not talk about single modality It measures, we will be looking at multiple epigenomic levels. Combined with genetics, we’ll be looking at the interaction of those probably at single cell level, and probably across both space and time. And we’ll be able to use that to predict disease before symptoms occur. And treat and monitor the response to that treatment in real time. In available fluids, hopefully, blood saliva, urine, things that are non invasive to the patient. That’s what I’ve been working towards my whole career, I think most precision medicine and near the data is gathering that we can do this, I think the tools are the molecular biology tools, the increase the decrease in the cost of sequencing, and the analytics, the analytics are super important. With the advent of really applied machine learning, the ability to integrate all these ohmic layers, and learn what predicts patient directories has come on incredibly in the last few years. And I think it’s going to transform how we do healthcare.

Steve Caldicott 26:18
Well, that’s a pretty good way to end, I think thank you

Jo Holbrook 26:21
for the invite, I really enjoyed talking to you. I’m really looking forward to the conference in Brussels in a couple of weeks. It’s an incredible programme. And it’s a exciting time to be talking about these things. Because, as we were talking about it’s becoming Prime Time,

Steve Caldicott 26:35
Cambridge, epigenetics, our platinum sponsor of our epigenetics and epigenetics in health care and disease conference in Brussels, 17th and 19th of November. You’ll see at the end of this, there’s a link to register or to view the agenda and we’re heavy, Joe, you’re going to be giving a talk at the events. And obviously Joe and some some of the other guys from the team at Cambridge, epigenetics will be on hand to answer any questions, or any ideas about collaborations and partnerships and what have you in Brussels. But obviously, aside from that, you know, if you can’t make it to Brussels, check out the website. We’ll put that in the link as well. And Joe is but I feel enlightened.

Jo Holbrook 27:25
I’m glad to hear that. I can go on about this stuff forever. So we talked about it. And thank you. We are really open to collaboration. We have an early access with our products. We’re also hiring. So please do check out our website for competitions. Going fast.

Steve Caldicott 27:44
This is lovely to meet you. And I see actually in person in Brussels that we can

Voiceover 27:49
see each other in 3d again. Yes, see that? That was precision medicine Forum Podcast. Visit precision medicine to get all the show resources and find out about our upcoming episodes and events and please subscribe or follow on your podcast app, so you never miss an episode.

Cambridge Epigenetix is now biomodal