Founder Stories: Chris Wigley, CEO of Genomics England

This is the BioCentury Show. The BioCentury Show is brought to you by the Second BioCentury Bay Helix East-West Biopharma Summit. Join us October 2nd to 4th in Kendall Square to debate how to globalize biopharma innovation to benefit patients and achieve an ROI for investors.

Hello and welcome to the BioCentury Show. I am Simone Fishburne, editor-in-chief at BioCentury. Happy to be joined today by Chris Wigley, CEO of Genomics England. Chris played a prominent role during COVID He was seconded as the senior responsible officer for data to NHSX, which is a joint unit in the UK between NHS, government, and industry. That aims to advance digital systems and technology to patients. And that is really also, I think, at the heart of what Genomics England to some degree does. Prior to that, Chris was a CEO at Quantum Black, which was focused on machine learning and AI. So of course is quite at the center of where innovation and technologies are going today. Chris is also a trustee of the New Entrepreneurs Foundation and on the advisory board of deep tech VC Entrepreneur First. So Chris, I do want to start by talking about Genomics England. You know, I think that people outside of the UK may have kind of heard of it, but aren't exactly sure what it does or how it interplays with industry. Tell us about Genomics England and what is the offer for biopharma there?

For sure. Thanks, Simone. And it's great, great to be with you on the show. So the origins of Genomics England actually start with a human story.. And we like to really try and relate everything we do back to the humans, the individuals, the families that we're helping. The family in this case was the Cameron family, and David Cameron happened to be the Prime Minister of the United Kingdom at the time. They had a son, Ivan, who had a rare disease which is genetically driven called Ohtahara syndrome. And like lots of parents, They got smart on that, learned a lot about the genetics. In their case, it impacted whether they were gonna be trying to have more kids and so on. Ivan sadly died in his late childhood, but they were left with an enduring sense of mission. And from that, as part of the Olympic legacy of the UK, the 100,000 Genomes Project was born. And so this was in 2013, just after the Olympics. It was only 10 years ago that the first human genome had been sequenced at a cost of billions of dollars and thousands of human years of work. And so the idea of doing 100,000 was pretty audacious at that point. It wasn't clear. If you multiplied the then cost of doing a genome by 100,000, you got to a number that was way in excess of the funding. So it started life with an interesting challenge. Of course, as we all know, the cost of sequencing came down almost faster than any other technology in history, which which is extraordinary. And at the level of sequencing in partnership with the NHS, the 100,000 Genomes Project was completed in sort of late 2018, early 2019. Work continues with those participants. What we've done to build on that sort of foundation for Genomics England is, again, in partnership with the NHS, we now offer whole genome sequencing for rare disease and cancer across England. It's the first country in the world to have nationwide whole genome sequencing for diagnostics. And so both the 100,000 Genomes Project, the work that we did through COVID, the clinical service that we now offer, all of these act as kind of engines of growth for data. And then in terms of the interface with research, we can then de-identify all of that data, make it available in our trusted research environment. And that is the genomes of the people, is their life course clinical data, If you've had cancer, it's also the tumor of your cancer, as well as your germline genome. We're increasingly adding high-definition pathology images, transcriptomics, proteomics, metabolomics. We're sort of enriching that dataset the whole time. And then the researchers, both in academia, startups, biotech, big pharma, do their work, drug discovery, diagnostic development, fundamental science. And critically, we can feed those insights back into the clinic. So we have that kind of live wire reconnecting us with the clinical service so that those insights can kind of be brought to bear for patients. So I can talk a bit more about how we work with biotech and pharma and so on, but that's the basic picture.

Right. So, you know, what I'm understanding is that, you know, it is Genomics England, but actually you have multiple omics that are in your purview or in your scope, let's say. I do think that I think there's one critical question because I think a lot of people, you know, there are actually many databases that are being gathered. And, you know, I think a broad question is how do you make sure that these databases are actually furnishing and enabling innovation rather than just becoming data collections that are sort of an end goal in themselves?

Great, great question. I mean, on the name Genomics England, when I first arrived here, I said, look, there's only two problems with the name Genomics England. One is the word genomics, and the other is the word England. Because genomics suggests that we only care about DNA, and England suggests that we only care about England. And neither of those things are true. One of our board members, Ewan Birney, has this great Latin phrase, "latu sensu," which means in the sort of widest possible sense, rather than "latu strictu," which is like in the strict sense, which we would say in English. And so he talks about it. Genomics Lato Sensu, meaning sort of all omics.

So—

but lots of people use it in different ways. Absolutely very much focused on the broadest interpretation of how can we use effectively molecular insights to inform both research and development and also clinical care. The question on datasets and kind of value is really important. Coming from a tech background, my personal bias is very much towards who is the user, what are they trying to do, how do we help them do that thing, rather than let's build a sort of perfect, you know, crystallized kind of taxonomy of all of the information in the world and kind of polish it like librarians. You know, that's, that's not what we do. So we, we tend to work back from, um, what are our users trying to do, what skills and capabilities do they have, And then try and continue to develop the asset in order to meet those needs. So for instance, one of the reasons that we've gone back and sequenced all of the— sorry, scanned all of the pathology images from the cancer patients is that talking to users, they said, look, we love having the DNA, both of the person and of the tumor. We love having the clinical information like the cancer registry and progression-free survival and all of these good pieces. What we really want to add into that is both RNA and the imaging. And so we're working on the RNA. The imaging is now there for the 100,000. And we're working right now with researchers like the Mahmood lab in Harvard around building models to say, how much of an incremental lift does it give us to have features that we can derive from the images using machine vision alongside the features that we can derive from the genome? To build better predictive or sort of classification models about tumor signatures, treatment options, and so on. So we're constantly trying to enrich it in order to kind of serve our users rather than as an end in itself.

Can you walk me through— so let's say I am a researcher either in a university, say Cambridge, or in a company. I go to Genomics England, and we'll talk about this in a minute, by whatever mechanism I'm able to access your information, how am I able to use that in my own research? So what am I looking for? I'm looking for patterns, trends, and how accessible is the information for me to go and generate a novel insight from that I can maybe patent or use?

Totally. Um, so if I talk first about how do I as a researcher get access to this, um, Google Genomics England research, um, our website will hopefully come up. There's a, there's a research section on the website.

It does, I checked. Yes, you'll be pleased to hear Google works, right?

Yeah, it works. Um, and on the, on the, in the research section of the website, there is literally a button saying apply to be a researcher. Um, there are separate paths if you're a researcher in industry or if you're in academia. Um, you need to validate your, um, institution At this point, we've been doing this for a few years, so most of the major research institutions and most of the major pharma companies and biotechs are actually already registered with us as institutions, which makes it easier to— if you're a new person from that institution, there's a sort of streamlined process to get in. If not, we can just add your institution or your company to it. That's pretty straightforward. It just takes, you know, a few days more. Once you're in, we talk about it as a kind of aquarium model. You can come in and study the fish you can't take the fish away with you. Um, you know, so we're not a fish shop, we don't sell data, um, we're an aquarium, um, you come in and you sort of study the fish. And so what that means in practice is you log into AWS WorkSpaces, there's then a range of, um, tools within there. Typically the kind of flows that people go through are they're interested in a particular phenotypic presentation, like a, um, a particular clinical indication, let's say like epilepsy or neurological development or musculoskeletal or whatever. And they want to understand— or it might be something much more precise, like I want to look at patients who are under 25 and have Duchenne muscular dystrophy and specifically have this variant in their genome, because that's the niche population that I'm working with. Because we're a disease cohort, we're not— we don't reflect the population. We reflect sick people. We're actually massively over-enriched for people with diseases, unsurprisingly cancer, rare diseases.

Does it help me design stratification rather than identify individual patients?

Um, so you can look at individual level, uh, patient data. It's anonymized, so you don't see their sort of name and address or whatever.

Right, right. So I can't necessarily recruit using this?

Um, well, good, great question. So let's follow this sort of workflow through. So let's say, right, I'm interested in a rare disease. There's 6 people in the database who have that. I can look at all of their data at an individual level. I can look at the individual genomes. I can look at the individual phenotypic presentations. But I don't know who they are. I can then put in a request to say, ooh, we're doing a clinical trial, or we're doing a study about this condition. We would love to recruit these people. Can you please recontact them and offer them the opportunity to take part in this research? Yes, we can absolutely do that. The process is not as streamlined as I would like it to be. I want to be clear, this is not like on a Thursday afternoon you say, I want these 6 people for my clinical trial, and on the Friday morning, you know, they arrive with a bow tie, you know, um, ready for that.

But because all patients wear bow ties and drug development is generally that efficient, so I think we'll give it a little bit of a—

I was probably mixing my metaphors there. I was thinking of a present with a sort of bow on the top.

I see, I see, rather than the—

okay. But the processes are all there, the consents are all there, all of them have consented to be recontacted, all of them—

I see, so it's pre-consented, right?

Yeah, pre-consented, and also there is a very strong incentive for people to take part in research because they're sick or their kid is sick or, you know, they're looking for cures, they're looking for new therapeutics, they're looking to improve life for themselves and also often the community of people with those conditions that they are part of. It's a very incentivized population.

One more question on this thread. I think a lot of people, even in the US where I live, in industry are employing UK Biobank. I know that many of those are in SMEs actually, not just the big pharmas. Maybe you can just explain a little bit about whether, first of all, it sounds like you are accessible to SMEs, but you could talk about that. But also what is your relationship, if any, and how do these structures, UK Biobank and Genomics England and so on, sort of fit together?

Great question. So let me talk about 3 institutions first, and then I'll talk about how we sort of fit together. So UK Biobank, 500,000 volunteers kind of early to mid-noughties was when they were recruited, so kind of about 20 years ago, typically in kind of middle age and broadly reflective of the population. Um, probably, you know, and Rory, who heads up Biobank, would say this, you know, they would love to have had more participation from underrepresented communities, but broadly it's a population cohort. Um, those people have now aged 15 to 20 years depending on when they were recruited, so it's increasingly very relevant for kind of later years' study as well around things like neurodegenerative and so on. Population cohort, anonymized, fully consented, and again continually enriched. So they've gone back to the samples over time and done all the exome sequencing. They're just in the process now of doing all of the genome sequencing. They're adding imaging, other things. So that's very much a sort of live data set as well, but population volunteers. Genomics England, sick people, often sick kids with genetic root causes for their conditions. So Biobank, typically middle-aged to later years. GEL, typically kids and young people and/or people with cancer. But sick people. Similar structure of continually enriching the datasets. And we've talked about how we interface. The third institution I'd call out is Our Future Health. So they're right now in the recruiting phase, have partnered with a bunch of the biotech and pharma companies. They're shooting to recruit 5 million people. So incredibly ambitious set of targets there. And their focus is really on predictive and preventative use cases. So things like polygenic risk scores, things like cardiovascular health, linking through to things like which conditions should we be screening for when at a population level, um, what, what are useful pieces of information to hand back to participants so that they can manage themselves better over the course of their lives. So those three organizations really sort of sit alongside each other, and we all collaborate quite closely. So for instance, with Biobank there was a, there was a modest overlap, um, in Biobank of— between Biobank and Genomics England of people who volunteered for, um, Biobank and then got sick, and so therefore were sequenced for one reason, principally for cancer. Um, and so we've identified, while working with the participants on both sides, um, the group that is in the overlap. Um, and actually we can then bring together all of the datasets from Biobank, all of the datasets from Genomics England for those individuals who've had cancer. Um, and it becomes a kind of 360-degree, 720-degree, whatever picture of, of those people. So incredibly kind of rich, um, research resource. So we're, we're pulling that together kind of as I speak. Um, but that's an example of the way that we're trying to increase the connectivity between, um, the different Thanks.

So we're going to take a brief break. And when we come back, we're going to talk about some broader issues for genomics and the situation in the UK.

Emerging biopharma companies in the US, Europe, and Asia face the same dilemma. How do you globalize to deliver innovation to patients and reward investors? A handful of biopharma companies will build global operations. Most will achieve global economics through partnering and licensing. Others will capture a share of the economics by enabling global development and financings. This October in Kendall Square, BioCentury and Bayhelix once again bring together decision makers and investors to address these industry challenges and to form trusted cross-border relationships. Register today at biocentrieastwest.com.

Back with Chris Wigley from Genomics England on the BioCentury Show. Chris, I want to talk about the status of genomics in the UK. Sitting in the US, and you know, you had a role obviously during COVID as we talked about at the top of the program, you know, sitting in the US, it really seemed like the UK and Israel in fact were real leaders in the pandemic in terms of genomics and their ability to use that information to track the evolution of the virus. So what did they do right, and how can the UK and maybe Israel as well sustain or leverage that advantage sort of beyond the pandemic?

Great question. And I think that in both cases, it's notable that both the UK and Israel have kind of universal health coverage and, you know, life course data and so on, and also that both countries have really invested over a long period of time in kind of the research base, um, and in capabilities like this. I think it's, um, when something happens suddenly, it's difficult to kind of build something from scratch. So at the time, there was a lot of debate about actually the UK is not great in terms of having a diagnostics base compared to, say, Germany or whatever. But I think the strengths that we had, one is an integrated healthcare system, single-payer There are obviously lots of subdivisions across the different bits of the NHS, but it's ultimately a single system. Someone has an NHS number, they're in the system. The second is the kind of longstanding strength of the academic base in the UK. And for some reason, from Darwin through Rosalind Franklin, Crick and Watson, Sanger, this has been an area where the UK has been really strong for kind of 150 years. And so we were playing a lot on that heritage.. But what that means is we had a lot of the infrastructure in place to be able to move quickly when we had to. So standing up the sequencing at places like Sanger, standing up the laboratories that were put up very quickly, the fact that we had invested in genome sequencing in the healthcare system already meant that there were regional labs that had Illumina machines, were good to go with all of that, all of those pieces. And I think the third angle is also that interface between research and the clinic meant that leaders like Martin Landray, others, were able to stand up things like the recovery trials very quickly, link that through to things like GP data, understand which kinds of therapeutics might be interesting for repurposing. And so when you get all of those bits of the ecosystem working together, We at Genomics England were looking at sequencing the humans, understanding what kind of glitches in our DNA made us more susceptible to getting severe COVID. Sanger and others, the whole kind of COG-UK coalition, were sequencing the pathogens, mapping all of those. The recovery trials were in the clinic. And then you had people like Patrick Vallance sitting like the sort of spider in the middle of the web, pulling all of these pieces together and saying, okay, these 2 or 3 things are an absolute national priority. Everyone has to do those, drop everything else and just do those. Um, and so I think that sense of shared purpose, clarity of priorities, and a kind of infrastructure base that we could draw on were absolutely essential. And coming out of the pandemic, I think the question is, how do we sustain that sense of mission? How do we sustain that sense of collaboration? How do we sustain that sense of really clear prioritization? As we work on some of these other challenges like cancer, like cardiovascular disease, like neurodegenerative, like obesity, these big medical missions that were laid out in the Life Sciences Vision of 2021, we need to bring that sense of mission to those goals.

So I'm going to spend a little bit of time now on what is difficult and what the real challenges are. I think we can probably agree that one of the most burning issues issues that people are becoming increasingly aware of is the databases are heavily skewed to white European ancestry. Some attempts to address it have actually made it worse. Maybe, you know, a couple of questions related to this. One is maybe you can just briefly outline efforts in and beyond Genomics England across the field. You know, what works and what doesn't? And I think it's also important to think about the fact that if drug development decisions are being made based on this data, are we actually at risk of exacerbating disparities? Because we're using, you know, unrepresentative datasets to make decisions. So perhaps you can talk a little bit about both of those things.

For sure. So at a technical level, this is very clearly a very real risk that we need to address. And so, I mean, our mission statement at Genomics England is bringing the benefits of genomic medicine to everyone. And we always kind of underline the word everyone. And we can only deliver on that mission if that's really true, if we are able to serve all of the communities that, um, you know, are in this great nation. Um, I'm coming to Genomics England from a kind of machine learning background. There were a bunch of examples of this in the, um, you know, in the development and rollout of machine learning. Things like facial recognition— if you train the model on a bunch of photos of white men, guess what? It performs more poorly if you're a person of color, if you're a woman, if you're older, if you're younger, etc. Um, and so we know that overfitting of these models is, is a real issue that we need to address. Now, how does that translate into genomics? Um, it's complicated, is the short answer. Um, it's— there is not a causal relationship between the box that an individual human might tick, um, of what ethnicity they self-identify as, um, in let's say a census survey, um, and their DNA, you know, those two things have some kind of relationship, but it's not a direct and simple relationship. Um, you can then get into, okay, um, deriving someone's ancestral heritage on the basis of their DNA. We can do that. We can, we can kind of cluster people into, for instance, um, content— continental ancestry groups. Um, but actually often The, the meaningful differences, um, lie in a much more granular, um, level of analysis. So for instance, what we're really looking for from our perspective, the, the problem we have to solve is, are the pipelines that we're running for the patients that we're serving, um, accurate? And so one of the first questions in the pipeline, I mean, in a non-technical sense, is, um, here, here are all these variants —compared to the reference genome, let's say 5, 6 million variants per person. Are those variants rare or not? And that's kind of one of the first points in the decision tree of the analytical pipelines. And variants that are rare in European ancestry folks are often not rare if you're from, say, the Xhosa population in Africa, right? And so we need to have a much better sense of— making, making sure we're making good calls, um, because the right question there is, are they rare for you? Um, and so I think a lot of, a lot of really good work has been done on the Human Pangenome Project, which has been published recently. Um, so we're going, we're going from one reference genome that you're assessing rareness, um, against to— it's about 40-ish now, looking to get to about 350. That will make that question much better. We at Genomics England, um, pitched for and got £25 million of funding to work on specifically this problem around diverse data. And I think your question about what works, what doesn't work, one of the— you know, we've engaged with lots of communities for a long period of time through groups like our participant panel and others, and we've learned that we need to start with the ear before we get to the mouth. Actually, a lot of these communities that are underrepresented in in biomedical research are justifiably skeptical, because a lot of stuff historically has been done which is uncool. And so we need to acknowledge that. We need to understand the implications of that. And we need to work with those communities on understanding how we can rebuild trust, rather than rocking up and saying, hey, we're really nice people and we're really great, so just trust us. It's all going to be fine. We need to listen before we speak. And then we need to work with leaders in their communities to engage in the right way. To be meaningful and build that trust. And then once we can do that, and once there's a sort of value proposition— so for instance, diseases like sickle cell disproportionately affect people from Black British, you know, African ancestry communities— we've partnered with the Sickle Cell Society to help try and solve that problem, right, of getting more equitable access to treatments for that terrible disease. And so we have to be clear about this is not just us being the sort of data vacuum kind of sucking everything in, but us working in partnership with those communities to understand what's important to them and work with them to try and, you know, do the right things to make sure that we're serving them as well as we can.

Right. And actually, that brings me to my next question, because as you pointed out, it's not really only about the biological question of the genome's ancestry. And that, of course, doesn't necessarily track with how somebody checks a box. It's also important to engage with patients and participants in multiple communities in order for them to be stakeholders. And so that brings me to the, you know, another really big hurdle across the field, which is the need for longitudinal studies. And, you know, that means getting participants to commit and remain engaged over the long term. I mean, Framingham maybe was the study that did this arguably the most successful. I don't know if there are lessons from there, but that was very contained. And so are there, you know, one question is how do you go about that? And another thing I think maybe you could feed into the answer is there are certain technologies that might reduce the friction, you would think, like wearables, you know, that means you don't necessarily have to go in. Are these game changers or can you talk a little bit about why we should actually believe that we can engage large communities or large numbers of people over the long term to do, to contribute this information?

Yeah, it's a, it's a great question. I mean, from a, from a Genomics England point of view, we're right now launching what is to some extent the mother of all longitudinal, uh, studies, um, which is our newborns work. So we've worked really closely with, um, mums, um, new parents, midwives, others for, for, and ethicists and scientists for many years to kind of work through some of the issues here around the ethics and practicalities of newborn sequencing. We got to a place where we were confident to launch a major research pilot of 100,000 babies. The first babies are going to be sequenced later on this year. That'll be 100,000 babies who are born with their DNA on file. And they're going to live 100 years. And so if I'm in flippant mode, I would say that's great, because when Treasury asks me for the one of the success factors and metrics for the study. I can say, ask me in 100 years. But it's actually a serious point that we need to figure out how to remain engaged with that community over the course of their lives, right? Because question number one for the study is, can we, rather than testing for 9 things using mass spec on the sort of heel prick test, can we test for 300 genetically driven which affect baby— affect young kids, you know, before they're 5 or 8. Um, we think the answer there is yes. You know, we want to prove that out through the— through the pilot. Second question is, okay, we're only passing back information at the moment on conditions which are treatable, because otherwise we, you know, through engaging with parents, they're like, don't tell us about stuff we can't do anything about, we don't want to know, right? Just tell us about stuff we can do something about. But we want to make more and more of those conditions treatable over time. So there's a big research piece there about small n and n-of-1 therapeutics. You know, how do we— how do we do that? So we've We've partnered with people like Julia Visorello, the mother of Mila, of Milasen fame, the sort of first N-of-1 therapeutic. But then the third question is, you know, how does this benefit people over the course of their lives? You know, in childhood, if you then turn up in A&E with an undiagnosed condition, does it help to have your DNA on file? We think so. We want to prove that out. Teenage years, you know, obesity, mental health, hereditary diabetes, middle age, you know, cancer, cardiovascular, later years, neurodegenerative, and so on. There is that whole life course relationship with the genome, which which we're really excited about. And I think that in the UK, for whatever reason, a bit like 150 years' work on genomics, we've actually done a lot of work on these long-term cohorts. So UK Biobank, which we talked about earlier, is a good example of that. Most of the cohort were recruited 20 years ago. Today, when Rory and team reach out to them and say, would you mind possibly coming in, spending a whole day being imaged, out of the goodness of your hearts. Um, lots of them have to drive an hour, 2 hours to get to one of these imaging centers. They get something like 60%, um, saying yes. And I'm making up the number, maybe it's 50, maybe it's 70, but, um, it's of that kind of order of magnitude. I think that's amazing, um, that purely out of the goodness of their hearts, 20 years after being recruited, people are still willing to hop in their car and drive a couple of hours and spend an entire day getting imaged, but they don't see the images, they don't get any benefit themselves. It's purely for the good of science. And as we've been looking at the newborns program, um, previ— like earlier efforts for these kind of longitudinal studies, one's called ALSPAC, and I can never remember the name. It's like Avon and something lifetime cohort. Um, that's maybe 20 years ago now. They're still engaged with that community. There's a great program called Born in Bradford, which is working with the community in Bradford. Um, the, the people who took part in that study signed up like 14, 15 years ago. Um, and so to some extent, groups like that can give us a truth set for newborns because we can say, okay, we, we think that these things are going to happen to this newborn baby. Did that turn out with these people who are now 20? You know, if we, if we looked at the same variants in those people who are now 20, because, you know, the DNA hasn't changed that much, maybe a bit of methylation here and there between, you know, now and then. Um, was that right or not? So I do think there's an incredible power in having that longitudinal relationship, but it's not something that comes for free, right? I mean, you have to really work at it. And I think Biobank, AusPac, Born in Bradford, others, you know, East London Genes and Health is another good example. There are tons. Um, and, you know, Genomics England is just coming up to its 10th anniversary July this year, we're going to be having a big party to celebrate it. You should come. Um, and a lot of the kids who were sick, kids who took part in the 100,000 Genomes Project, are now in their early teens. And actually, once they're 16, we need to recontact them. They need to choose for their data to be retained by us, um, in the same way that, you know, the biobank participants need to choose to hop in their car and drive. We've been saying, without levity, should we have a Genomics England TikTok account to kind of, you know, be in the flow? These kind of 13 and 14-year-olds, right? It's a real thing. You know, how do we communicate with them? Just like running town halls about newborn sequencing, we have to go to the people. We can't expect the people to come to us. And for 13, 14-year-olds— I mean, I have two of them myself— you know, they're not going to read a letter that comes through the front door. You know, they're not going to We need to engage with them in a way that they want to be engaged with.

Well, I can tell you that in my family, I've been informed that when they say, "I read an article," it means they saw a TikTok. So I think you're on the right track there. Chris, I think that's a great place to end. We're out of time. This has been a fascinating conversation. I will not wait 100 years to follow up with you. I'll be looking to see the progress of Genomics England and genomics writ large, actually, and omics writ large. Continue to follow that at BioCentury. So thank you very much. And it's been a great conversation. Thank you.

Thanks a lot. The BioCentury Show is brought to you by the 2nd BioCentury Bay Helix East-West Biopharma Summit. Join us October 2nd to 4th in Kendall Square to debate how to globalize biopharma innovation to benefit patients and achieve an ROI for investors.

Mm-hmm.