Browse Our NASH and COVID-19 Coverage
Matt Kelly, Chief Innovation Officer at Perspectum, Ltd. joins the Surfers to discuss two studies that look at the relationship between COVID-19 and the liver. In the COVERSCAN study, Perspectum plans to conduct multi-organ scans of 500 patients who have recovered from a COVID-19 infection at two-time points: post-infection and six months later. The goal was to establish the impact of the COVID-19 infection on all the body’s major organ systems. During this podcast, Matt discusses the results of the first round of scans on the first 200 patients. In the current UK BioBank analysis, UK BioBank culled through its database of 500,000 patients aged 40-75 to find 1,000 who had prior COVID-19 tests and imaging and compared the 10% of these patients who tested positive to the rest of the cohort. Both sets of results suggest a complicated relationship between fatty liver and severity of COVID-19 infection. This new research is producing eye-opening results that challenge conventional thinking. Perhaps our most important episode to date…
Roger Green:For everyone with an interest in NASH, or more broadly in fatty liver disease, surf’s up. Episode 29 of Surfing the NASH Tsunami starts now.
Roger Green: I’m out of quarantine today, back at the place I normally broadcast from, thereby marking the end of my travel log openings, at least until the next time I’m silly enough to get on a plane and go somewhere. I know you’re not going to miss that.
Roger Green: This is a special week for us. From time to time, literally since the first episode and why we started this podcast, we’ve talked about COVID and alluded to the idea that, while other end organs relate to COVID only in terms of outcomes, the liver can play a unique role in terms of actually being able to predict severity of COVID infection if NAFLD is present.
Roger Green: I’ve been waiting to do this episode for a while. Two things in the last two weeks have made it a really good idea. The more obvious one, if you live in the states or pay attention to what goes on here, has been what happened to our President, First Lady and a whole bunch of the White House staff in the last week. The less obvious, maybe more important one, though, was the initial release of results from the COVERSCAN study two weeks ago. COVERSCAN being a large study taking a look at COVID and its impact in relationship to all the major end organs in the body, sponsored by Perspectum Ltd. of, Oxford in the UK.
Roger Green: And today, we are exceptionally fortunate to have Matt Kelly, the Chief Innovation Officer of Perspectum, with us to discuss some of the early findings. The rest of us will ask questions and talk things through as we go along.
Roger Green: Matt, do us a favor. Take a couple of minutes; tell our listeners a bit about your background, your work and other points of interest in your life.
Matt Kelly: Yeah, of course. So nice to meet you all. I’m Matt Kelly, I’ve been at Perspectum for nearly five years now, and my role there is the Chief Innovation Officer. And the Innovation team, what we’re responsible for in the company is really the generation and dissemination of clinical evidence to support our technology and products. We do that through executing clinical studies, so COVERSCAN is one of those studies, and also clinical collaborations with key opinion leaders and clinical experts worldwide. We also secure and run a large number of grant-funded projects, and that helps us fund the development of technology and also to validate it.
Matt Kelly: I started my career, a while ago now, as a neuroscientist, before deciding to pursue a PhD in a Cambridge spin-out company, looking at computational drug design. And that was when I realized, after spending a year at a drug company, that life in a wet lab really didn’t suit me, so I was quite keen to find a way of working with a computer, to run the ‘n’ on the experiments and take it out of my hands.
Matt Kelly: So I enjoyed the PhD and after a brief stint working as a software engineer, I joined the University of Oxford as a postdoc where I focused on clinical decision support and was also introduced to the world of medical imaging, in particular looking at liver cancer. Then that led me to my next role, which was at Siemens Molecular Imaging. So moving from a small company to a big old academic institution, now to a big old company. There, I was the science lead for oncology, working mainly with pet CT imaging. But after eight years at Siemens, I realized I was missing some of the excitement that you can only really get with early-stage companies.
Matt Kelly: So then I joined a small, but growing company, Perspectum. There were about twenty employees at the time, but over the past five years, it’s grown to around about two-hundred people now with offices still in the UK and Oxford, but also San Francisco and Dallas, and then Singapore. It was Perspectum really that first introduced me to the huge burden of fatty liver disease. And if I’m perfectly honest, before I joined, I really hadn’t heard about it. To me, that highlighted that this need for activities such as this podcast and all the work that’s been done through the Global Liver Institute to really raise awareness of these conditions, because it’s, I think, obviously within the liver community, it’s well-known. And it’s increasingly well-known outside, and we’ve seen, with COVID, and I’m sure we’ll get to that later, some of these recent reports linking preexisting diseases such as metabolic disease and fatty liver disease, to how well you cope with COVID.
Matt Kelly: But yeah, I’ve enjoyed working in this field for the past five years. It’s an exciting field, there’s lots of opportunity. There’s lots to be done. It’s an important field to work in.
Roger Green: Thanks for that. Matt also shared with us before the podcast that his kids are in Tae Kwon Do, but it’s socially distant, no contact Tae Kwon Do, which is one of these concepts that I’ll be so happy for the end of 2020 because some of those might actually go away. I’m trying to visualize what no contact, socially distant Tae Kwon Do looks like. If we have time at the end, maybe you can tell us a little bit.
Roger Green: Thanks for the introduction, the things we do before we get started. I said last week, we hadn’t seen our Episode 27 numbers; they’re virtually identical to our Episode 26 and 25 numbers, which is great. We said we were going to talk a little bit about digital AASLD, known as TLMdX, this week; I want to get this topic maximum time, but please note the meeting runs through November 13 to 16. We will be looking to record some kind of event on each the pivotal days and have the sessions ready before people in the Eastern US wake up the following morning, and we will be announcing times of things, preview episodes, KOLs and all that stuff starting next week.
Roger Green: Thanks to those of you who called or emailed me suggestions. And if you want to stay on top of what we’re doing, please make sure go to the Surfing NASH page, give us your email address. We can add you to the growing list of people who get updates from us every week.
Roger Green: One of the things we’ve done in the last week in anticipation of today’s… Actually, originally we planned to do this after the president got sick, but certainly with today’s podcast, is we now have a COVID page on the Surfing NASH portion of the HepDy website, which gathers all the episodes we’ve ever had that have discussed COVID. We put them all in a single place and you’ll find this one there as well.
Roger Green: Because of everything else we’re doing, we’re going to take a hiatus from music this week while we scour for more sources of liver music, arranging, as we have in the past, from the exotic, that would be the Cirrhosis Song, over to the magnificent, Jeff Ramsdale’s work or Naim’s. At any rate, this week we’re going to be silent at the end, or at least we’re not going to be musical at the end.
Roger Green: And with that, onto our podcast. I like what we did last week; people can do a personal or a professional best thing of the week, whichever one you want to cover. Brave person go first, let’s go.
Donna Cryer: My first professional success is getting this microphone to work. I believe I can say we’ll see at the end of the episode.
Donna Cryer: But personally, I was able to have a socially distant walk with two friends yesterday. That moment, or two hours of social connection that I have not had for so many months was truly magical and healing. So I’m very grateful for it.
Roger Green: Excellent.
Stephen Harrison: This is Stephen. So for me, just more on a personal note, my son is a sophomore at Texas A&M and came down with COVID about three weeks ago; had a relatively rough course for a twenty year old. He didn’t have to be admitted to the hospital, but he had a bit of a go of it. Dad made several trips, throwing him medicine into his apartment without trying to catch the virus myself. After he recovered and he sat through his quarantine in his room, we actually connected with him this weekend. He was able to come home and we had some parent-son time together. It was good just to reacquaint with him and know that he was well. So since we’re talking about COVID today, I thought I would mention that.
Donna Cryer: First, full recovery?
Stephen Harrison: Yes. fully recovered.
Donna Cryer: Good.
Roger Green: Excellent.
Louise Campbell: Excellent news. I’ve got only personal this week and I’m looking forward to having a brand new kitchen installed. It’s taking a lot of time, but it will get there. Nothing exciting, but something that’s going to be very pleasurable and good to manage, and cooking.
Matt Kelly: I’d say from my side, it’s probably obviously the excitement of the boys being back at school and helping them with their homework. So my eldest, Fraser, was making a 3-D model of a cell, of a eukaryotic cell, over the weekend. So that’s been good fun, and refreshing my basic biology, which has been always good.
Roger Green: That’s fantastic. Okay, so I have two, and actually they’re both personal. One is, to all my local friends who thought I was out of my mind to go to Corpus Christi to visit our granddaughter, I’m now out of quarantine. So, it’s possible, I think, to get in and out of Texas and stay negative, as Stephen has been able to do successfully for the last several months while living there.
Roger Green: The second one, and this is something only Louise can fully appreciate, my favorite football team, Tottenham Hotspur, had a spectacular result yesterday, their best ever win over Man United. If you know soccer, then you know that historically there are… football, pardon me. Two kinds of fans, Man United fans and people who who root against them.
Roger Green: Of course, on the day we chose to do that, Louise’s favorite team, Liverpool, suffered such a devastating defeat that we were not even in the top story on the sports pages. Louise, I want to thank you for that, but we take our victories where we can, and, come on you Spurs.
Roger Green: All right. With that, and Matt, I never asked you who your football team was, but we can get to that later as well. It’s kind of a hazing ritual on this podcast, off-air. Okay, again, as I said, we’ve been talking about COVID forever; three of our first four episodes are about COVID, one about clinical trials, one about commercial strategies and one about how to forecast the impact of COVID in the long run in liver disease.
Roger Green: We came up with this whole broader NAFLD concept based on the nascent idea at that point in time that there might be a relationship between liver disease and what it meant for the outcomes of viral pandemics or epidemics, and how we would deal with that. And we’ve talked about it probably in around a third of the episodes we’ve had so far in this podcast. However, most of that’s been highly speculative.
Roger Green: COVERSCAN, I guess the first webinar on it was September 22nd. I was delighted to be on that, and as we said, the next week, the President and much of his White House staff came down positive and it just felt like this was really a good time to talk about that issue since it’s been observed that a fair number of folks around the President, certainly, are obese. It’s been noted frequently that obesity has a connection to the outcome of a COVID infection, and age, but nothing else really has been talked about in the US media. So I thought this was really a good time to bring this subject up and educate our audience and maybe educate people who aren’t usually in our audience.
Roger Green: Matt, why don’t you please kick us off. Take maybe six to eight minutes, tell us about COVERSCAN and the trial and some of the results. If you go a little bit longer, that’s fine.
Matt Kelly: Of course, happy to. So, as you mentioned and introduced, one of the things that we did when the pandemic started, apart from physically closing our office quite early to prevent any spread within the company, was look at what we could do to support. At the time, we had some technology that we’d been developing with Innovate UK grant funding, and that was to really look at multiorgan morbidity in diabetic patients. So we had this technology, it was in development. What we were looking to do is look at, how can we apply this? Because the early data from the UK and from Asia, and also from the US, was showing that comorbidities or preexisting chronic health conditions seem to be playing a major role in the severity of the infection and also mortality associated with it. So we thought, well, this is a good opportunity for us to really characterize: what is the level of organ morbidity in patients who’ve had COVID?
Matt Kelly: So we set up this study called COVERSCAN. We went through the ethics procedure; this was greatly accelerated in the UK. They were really doing an extremely good job at any COVID-related studies, at making sure they got a rapid accelerated review. I think from starting the study to getting ethical approval was under a month, or starting the concept of the study, so record-breaking time for us, and we set the study going.
Matt Kelly: So the idea was to recruit just over five-hundred patients who’ve had COVID that are no longer thought to be infective, so they don’t have an active fever, persistent cough, or any loss of taste or smell at time. And then we would scan them once with this multiorgan scanning procedure, and then again six months later to see if there’s any recovery or any change in organ pathology over that time.
Matt Kelly: The intention is really to inform healthcare providers of what the longer-term impact is of COVID on healthcare services; where do resources need to be directed, what kind of organ involvement are we seeing in COVID? So patients come, they undergo a questionnaire where we understand more about their medical history, the severity of their disease, and also put them in the scanner. Then we’re scanning in two sites in this study at the moment. One is in the Mayo Clinic in London, but we’ve also got a mobile MRI scanner, which is basically, a truck turns up into our car park and our main office, turns into a bit of a Transporter and the size expands, things move around. And then there’s an MRI scanner in there.
Matt Kelly: So, we’ve been scanning patients at our facility. We’ve converted some of our offices into a mini clinic so we can take blood, et cetera. And then they go out and get scanned in this mobile scanner. The recruitment has been phenomenal for it, obviously because of the awareness of COVID and the sort of infection rates. Loads of patients want to take part, so recruitment’s been great and we’re expecting to hit our target well ahead of time. As I mentioned, we’re using this technology we developed; we call it Atlas internally. This is a multiorgan imaging protocol that acquires and characterizes the state of multiple organ systems and looks at body composition in a scan that takes about forty to forty-five minutes. And then we’ve got an AI-driven image processing pipeline that we’ve developed internally, that then processes this data with the help of image analysts and clinicians to check the final results, to understand the clinical impact. This pipeline is working really well, and we’ve scanned, coming up to three-hundred patients now, and we’ve got the results through from over the first two-hundred. So we’re in the process of pulling these together for a manuscript submission very shortly.
Matt Kelly: What we found was, on this first two-hundred patients, is that about twenty percent of our recruited patients were hospitalized. Hospitalization wasn’t a requirement, but it more reflects what we were seeing in the community, and twenty-one percent were hospitalized. So, they’ve had symptoms that are sufficiently severe to require hospitalization. The main symptoms we found were, as you’d expect, fatigue in almost everyone, shortness of breath, muscle joint and headaches, and then cough and fever. But cough and fever were in a slightly smaller percentage, only about three quarters of-
Matt Kelly: … cough and fever in a slightly smaller percentage, only about three quarters of subjects. And on average, there was about three months between the onset of symptoms, and then being scanned for the study. That’s the kind of time gap that we’re looking at. And then, in terms of the results for the organs, we found, surprisingly, relatively, a very small percentage had an indication of reduced lung function three months post COVID, so only about three percent, based on the fractional area changed. The lungs seem to have largely recovered in our patient population, but we saw significant injury in the heart. Over a third of patients had reduced left ventricular ejection fraction, and this was more common in male subjects, and more common in patients who had been hospitalized, so it was associated with more severe disease.
Matt Kelly: For kidneys, only about five percent, six percent had any evidence of injury, so again, that was relatively small, but more common in obese patients, and those who are hospitalized. Moving on to the pancreas, about sixteen percent had evidence of inflammation in the pancreas through T1 mapping. Again, over a third had evidence of fat accumulation. But based on this data, and previous data that we’ve also published, looking at UK Biobank, which we can talk about later, we think that this pancreatic fat deposition is more an indicator of metabolic disease, which is a risk factor for COVID severity, rather than a consequence of any infection.
Matt Kelly: And then in the liver, again, we saw about a fifth of our patients had evidence of fatty liver disease, and this was in line with what we’ve seen in other population studies that we’ve done in the UK. But eight percent had evidence of fibro inflammation in the liver, which was above the population average. Again, whether that’s representative of being at increased risk of testing positive and having symptomatic disease, or is a consequence of the disease, possibly through this inflammatory cascade that we are seeing in a subset of COVID patients, we’re still to determine. And the second scan at six months time point will hopefully shed some more light onto that.
Matt Kelly: Overall, over half of patients that we scanned had evidence of injury in at least one organ, with about a fifth having evidence of multiorgan injury. And what we found was that the BMI and organ fat were risk factors for disease severity, so you mentioned BMI as a risk factor earlier. And what we really saw was it was the combination of BMI, with also fat deposition on the liver and the pancreas that was this key factor for disease severity. It does flag to us that if these results are upheld in the full study, that there are certain organ systems that are going to require increased attention in patients that have known to have COVID, whether that’s through screening, or investigations, or just keeping an eye out for any signs of degradation in those in the longer term. We’re really excited about the study, we’re thrilled at how well it’s gone. And we’re just excited to share the results more widely.
Roger Green: Fantastic. Let me invite my colleagues to ask questions first. Louise, you want to go ahead?
Louise Campbell: I thought it was an excellent piece, and thank you, Matt. That was great. What I did also notice when I looked at your data, is that it also, I think you comment in it that the ethnic mix was also supported, ethnicity, diversity was also supported with COVID?
Matt Kelly: Yeah. So in this study, and in particular in the UK BioBank study that we’ve also done, we have seen that nonwhite ethnicity is one of the factors that is associated with increased likelihood of testing positive for COVID, and having sufficiently severe disease that it requires hospitalization. So yes, age, BMI, ethnicity, and male gender are all things that are risk factors.
Louise Campbell: Did you have enough data yet to say whether or not they were also populations that had high NAFLD, because a lot of those populations that we know are genetically predisposed to developing NAFLD? So is there enough data yet to say whether or not that NAFLD in that population may be driving that ethnicity split, or whether or not it’s just not known yet?
Matt Kelly: Not in our data set yet, unfortunately. I guess one of the limitations about the study is that eighty-five percent of the population in cover scan today are white, so that doesn’t leave us with a huge number of non-white subjects to really drill into their contribution and how it correlates with, or is additive with liver fat. And the UK Biobank, the other study which has far more participants, 100,000 with imaging, again, even higher percentage of those are white, ethnically. So we don’t have a huge amount of data, unfortunately. And I think this is why we’re keen to run similar studies, or collaborate with people doing similar studies in other countries where there is a more mixed population.
Donna Cryer: Matt, this is Donna. First of all, I want to thank you for doing this work. Every day I get questions from liver patients who are asking for specific information about how COVID is affecting them, and what their risks are, and then what the course of disease looks like. And so I really appreciate you doing this type of research. One of the things that that struck me, and I have sort of as an ongoing question is, what this tells us about how and where research can and should be done in a post-COVID environment. So you used converted office space, it wasn’t a clinic, it wasn’t a traditional setting. You reached out to people in an untraditional setting, it was imaging based, not biopsy. So what does your study tell us about maybe the artificial limits and constraints that we had put on research prior to COVID, that are proving just that artificial? And what does it tell us about how we can do, particularly community-based research moving forward?
Matt Kelly: I think that’s an excellent question. And this cover scan study has certainly flagged a number of things for us that we wouldn’t necessarily have thought would have been so successful in the first place. So one of the motivations for using a mobile scanner in a nonclinical setting, so it’s on an industrial business park in Oxford is, as we talked about just before we started recording, a lot of patients aren’t comfortable going to hospitals at the moment, because of the perceived increased risk. And another issue we had seen with other studies in the past is, often there’s a resource capacity issue with scanning. So getting time on MRI scanners in some places, especially in the UK, can be tricky because it’s a heavily utilized resource and it’s not run round the clock. So we took the conscious decision to not run it in that hospital environment in Oxford, and to use a mobile scanner in our office area, in our car park. And that’s worked brilliantly, because we’ve not had issues with patients fearful of coming.
Matt Kelly: We’re able to control it from an infection risk point of view very well. None of the people that we have in perspective, who are running the scanners, or meeting the patients, or taking bloods have caught COVID, or had any kind of unnecessary exposure. We use PPE, there’s very little physical contact between people beyond the blood test. When you’re in the scanner, the patient can climb onto the scanner themselves. You have to put the coil plate, rest the coil on top of them, but there is really very minimal physical contact, and everyone’s wearing masks. So that’s helped. I mean, it’s shown us that, actually, if you need to scan a lot of people in a short space of time, then there are options to look beyond doing that in a clinical or a hospital environments.
Matt Kelly: I mean, the UK BioBank study, for example, also learnt this lesson a few years ago when they set up this study, which is scanning 100,000 people from the UK population. It’s scanned about 50,000-60,000 so far. And again, instead of trying to coordinate access to a hospital to scan patients, they bought an MRI scanner, rented an industrial unit in industrial states in various parts of the UK, and set up a factory, effectively, a scanning factory. And they get patients through, and volunteers through at such an incredible rate. And they are scanning fast. So Perspectum, we supply the liver imaging protocol and the post-processing for that study. And we were given something like three minutes of scan time to get all the data we need to characterize the liver in these populations. And they get it done, it’s so impressive. So I think that’s one thing, whereas, hospitals are set up very well to treat patients and to manage clinical management, et cetera. But if you’re looking running a clinical study or a clinical trial, then sometimes a more industrial approach could be beneficial.
Donna Cryer: Excellent. Thank you.
Stephen Harrison: Hey Matt, thanks for taking the time to give us an overview of the study. I have a couple of questions in follow up. Maybe the first one is an easy one. COVERSCAN: how did you come up with that name?
Matt Kelly: That’s an excellent question. I don’t know the answer, I’m afraid. I suspect because the first three letters are COVID, similar to COVID, but I honestly don’t know. I mean, some of the study names I’ve seen for other studies have been fairly sketchily linked to a phrase, but for cover scanner, I really don’t know the origin. I suspect it was our CEO, Banjo, who came up with a name in a moment of inspiration.
Stephen Harrison: Well, maybe it is COVID, somehow connected in with that name. Another question I had in looking at, you said you had scanned about two-hundred patients.
Matt Kelly: Yep.
Stephen Harrison: Twenty percent of them were hospitalized, which would tell me, if my math is right. That’s about forty patients.
Stephen Harrison: Can you distinguish anything unique about those forty hospitalized patients, compared to the others, relative to multiorgan disease liver severity? I want to ask you about corrected T1 in a third question, because I don’t think we went into that. But just top line, are you seeing anything different that you can comment on, relative to that twenty-percent that were hospitalized?
Matt Kelly: Yeah, and actually you flagged two of the main differences we saw. In hospitalized patients, they were far more likely to have inflammation in the liver, and evidence of inflammation in the pancreas. So those were two of the key , that we had. We also saw you were more likely to have kidney injury, as well. So the degree of damage or injury to these organs, this kidneys, the liver, and the pancreas was definitely correlated to whether or not the subject was hospitalized. We saw a slight increase in people with reduced cardiac function in the hospitalized subsets, but that wasn’t as extreme as we saw in the other organs. So the kidney, the pancreas and the liver, the pancreas and the liver in particular.
Stephen Harrison: Okay. And just kind of an immediate follow-up to that, in just thinking more about the cohort that were hospitalized, obviously, you didn’t scan the patient at the height of their illness. What was the mean time that they were scanned following discharge from the hospital?
Matt Kelly: It was about three months, the interval.
Stephen Harrison: So would it be fair to say they were probably, potentially even worse three months earlier?
Matt Kelly: I mean, symptomatically, certainly, they would have been in a worse state. I guess it’s all determining whether the organ injury, is how long that took to build up. Was it a sort of delayed inflammatory response, or something more acute? And I think when we get the data from the follow-up scan at six months, we’ll see how quickly it changes. Is it relatively stable, or has it improved? Or in some cases, hopefully not, it may have even got worse. But it’s an excellent question, and we’re really looking forward to finding out.
Stephen Harrison: Sure.That’s exciting. What about just moving on to CT-1, do you have the overall mean CT-1 for those hospitalized, compared to those that did not get hospitalized?
Matt Kelly: They were both in the normal range, so they were around, I’m trying to remember now. I think they were around about, sort of 760-770 in those that were hospitalized on average. So obviously, there were some lower, but there were quite a few that were outside that normal range of eight-hundred milliseconds, even above the higher threshold of 825, which predicts longer term clinical events. And for the non hospitalized, it was lower. It was more around 720, 730, with some still outside that normal range. But it was certainly significantly lower in the non hospitalized group.
Stephen Harrison: Okay. Yeah, I think that’s very, very, just incredible data. You mentioned early on when you started to, you did kind of a, it was a whole body scan, right?
Matt Kelly: Yeah, it was the thorax and the abdomen we looked at.
Stephen Harrison: Were you able to get any data on sarcopenia at all?
Matt Kelly: We have, we’ve collected that data. That’s not gone through our pipeline yet. So we’ve looked at body composition, in terms of subcutaneous, visceral fat, but also exactly that. A marker of sarcopenia, looking at muscle, because again, some of these studies are showing that sarcopenia is a huge risk factor for long-term outcomes, but also, if people have been hospitalized-
Matt Kelly: …it will be interesting to see what impact that may have had on their muscle mass.
Stephen Harrison: Absolutely. Well, it’s fascinating. I hope before we finish the podcast today, I can circle back around and ask you. We just touched on cT1 relative to what normal is, which is you guys define that as, what, less than 800 milliseconds. And that 825 has been predictive of long-term outcomes and data that you guys are generating in spades out of the UK Biobank. So, maybe before we end, you can give us an update on what that looks like. Because one of the interests that we all have is the context of use of a non-invasive test to predict long-term outcome, and you guys have been really the leader in pioneering what that might look like relative to MRI-based technology.
Roger Green: So, Stephen, back in the days when I did presentations for a living, we would give anybody a dollar to the charity of their choice if they could predict the next slide. You’re the winner for today because my question was going to be you’ve alluded to the Biobank work a couple of times. And I was wondering if you could talk about that first in the context of this study, and then the second, as Stephen pointed out, the context of what we’re learning about cT1.
Matt Kelly: Yeah, of course. So, the UK Biobank study was our first output really linked to COVID in terms of a paper. And I think we were really benefited by the fact that the UK Biobank existed. So, it’s been a hugely valuable resource. There’s a big UK population study where they are scanning. The study has enrolled at 500,000 people aged between forty and seventy-five. And those patients, they all get detailed questionnaires looking at their lifestyle, their diet. It’s linked to their medical records from there forwards. So, they’ll be followed up and we can see what diseases they develop, what interventions they have, and how, eventually, when they die and what they die of. But a subset of a hundred thousand were randomly selected to undergo comprehensive imaging phenotyping where they were scanned with an MRI of the brain, the heart, and the abdomen. So, again, look at how these imaging phenotypes link to genetics, blood-based biomarkers, lifestyle, et cetera, and follow them up.
Matt Kelly: So, it’s an amazing study. It’s been going for, I think, nearly ten years now. But this was going on and then COVID happened. So, it became an even more valuable because we had very detailed phenotyping and genotyping of all of these patients. And then a fairly big chunk of them will go on to have COVID. And especially because we’re looking at that age range forty-plus, you start to get into the higher risk categories. So, the UK Biobank worked in response round the clock to link testing data to the UK Biobank subjects and release that as quickly as they could.
Matt Kelly: So, in one of those early releases, we looked at it because we had results for a thousand subjects who’d had a COVID test and had also had imaging, and about a hundred of those, so ten percent tested positive for COVID. So, again, reasonable numbers, but we could start to see which factors that existed pre-COVID associated with the risk of not only testing positive, but also being hospitalized. And what we found was in agreement with what had been reported in other studies. So, being male, being non-Caucasian, you are more likely to test positive. Also, if you are from a lower socioeconomic background, again, you’re more likely to test positive. So, those factors weren’t surprising.
Matt Kelly: And then obviously our interest is in the liver imaging. We looked at obesity and we looked at liver fat. And what we saw was obesity on its own wasn’t a risk factor, wasn’t a significant risk factor to make you more likely to test positive or more likely to be hospitalized. And liver fat on its own also wasn’t. And it could be that it hadn’t reached significance in this n, but separately and independently they weren’t risk factors. But when you were obese and you had significant liver fat, so liver fat over ten percent, then that’s when we started to see these risks.
Matt Kelly: So, if you were obese with fatty liver disease, then you were more than twice as likely as the rest of the population to test positive for COVID, and you were more than three times as likely to be hospitalized with a positive test. So, that combination of those two was a real risk factor. And what we’ve seen in this and other studies is that actually only fifty percent of, or half of obese people, have completely normal liver fat in the UK in the Biobank dataset. So, it’s not a kind of nailed on fact that if you’re overweight, you’ve got fatty liver disease. And about twenty-five percent of obese patients had liver fat above ten percent so a quarter, really. And those are the people that would be most important to identify would be those obese patients who’ve also got a significant metabolic disease as evidenced by a fatty liver.
Roger Green: Okay. Stephen, did you have a follow up or information you wanted to ask about in context of cT1? I know you mentioned that a moment ago.
Stephen Harrison: Yeah. It’s just about any newer information. And quite frankly, it would be important maybe for Matt to review for the audience if he can, off the top of his head, what’s been generated relative to long-term outcome data. As far as I remember, there has been one early publication and then an abstract presentation. But do you have any more data that you could share with us, Matt, or maybe a publishing we might have missed?
Matt Kelly: Yeah, of course. So, actually, there was an update of that original outcome study that was published recently in “Liver International”. And just for everyone listening, there were about two-hundred patients. This was from Oxford University Hospital. Two-hundred patients with compensated liver disease. And they were all scanned at a baseline time point and followed up for a median of forty-two months follow up per participant. So in total, that’s just over seven-hundred patient years of data.
Matt Kelly: And what we looked at again there is how does baseline cT1, along with other baseline tests like biopsy and blood-based tests, how do they predict clinical events related to liver disease. So, we looked at all cause mortality, liver-related deaths, and also just the onset of some events. So, whether that’s decompensation, the development of asceites, varices, anything that’s classed as a significant clinical event related to the liver.
Matt Kelly: And what we found was the two best predictors of liver-related events were cT1, so this measure of fiber inflammation. And that was a threshold of 825, which for people like Steven who’ve been with perspective for a long, long time, equivalent to what used to be live to. So, this is our threshold that was specified in the original publication. And the other one that was equivalent to that was biopsy. So, looking at Ishak fibrosis from an invasive liver biopsy. So, those were the two that came out as the best standout predictors of outcomes. And there wasn’t really anything between them in terms of this prognostic ability.
Matt Kelly: The other tests that we looked at, we looked at FibroScan and FIB-4. They were also significant predictors, but not with the same performance as cT1 or biopsy. And also, they did an analysis, the authors at the university, looking at technical failure rates and found that, again, cT1 maintained its predictiveness. But the FibroScan there, it lost its significance when the failed scans were incorporated because there was a higher number of technical failures in the study. So, that’s our outcome data, our latest outcome data, that we have. And we were particularly excited about this because it was showing that biopsy derived fibrosis, which when we speak to drug companies who are running trials, we talked to regulatory bodies, that is really the nailed on, “okay, I want to know what the fibrosis stages”, because that’s prognostic. That’s what’s going to predict how my patient is going to progress.
Matt Kelly: And to find that cT1 was performing effectively equivalent to fibrosis was a really exciting result for us because that gives us confidence, more confidence, in its utility and this desire that lots of people have, and I think everyone on this call, to slowly move away from or quickly, even better, move beyond the biopsy and look at how various non-invasives, whether drugs or imaging, can be combined and put into a clinical diagnostic and prognostic pathway to identify the subset of patients who were most at risk of severe outcomes of their liver disease and require the most intervention as early as possible.
Donna Cryer: I swooned when you used the name of our campaign, I’m just going to say that.
Matt Kelly: I know.
Donna Cryer: Thank you.
Matt Kelly: It’s a great name, though. It’s a good name. Where did that name come from?
Donna Cryer: I invented it. When you have a lawyer who passes through a PR firm and on the way to a clinical trial recruitment firm, sometimes good things happen.
Matt Kelly: No, it’s a great name. It sticks in your head. You’re not going to forget it.
Donna Cryer: Thank you.
Roger Green: No, you’re not. You’re not. So, I had a question. On the day that you started doing this work back in March, April or whenever, were there hypotheses that you folks had in mind that have been borne out, or maybe more interestingly, things you thought you were going to find that you haven’t found, or things you found that you didn’t expect to find? Any of those three, really.
Matt Kelly: Just because of the early reports that were coming from clinical centers that had that early influx of COVID, we’d expected there to be involvement of liver and metabolic disease just because you were seeing diabetics at increased risk. A lot of the victims that you were seeing were overweight. So, that, to me and to us, felt expected. We didn’t know what the degree would be, we didn’t know how the interaction with obesity and fatty depositions on the organ would interact, but we thought there was something there.
Matt Kelly: I think we were all surprised at the degree of injury that we’re seeing in the heart in these subjects. I think that was certainly more than we’d expected. That wasn’t something that was particularly reported early on. So, that’s an exciting result and something that we’re certainly keen to investigate more. And one of the great things about MRI is there’s not really any better tool to characterize every component of the heart out there. So, there’s a whole suite of metrics that we can derive to really characterize cardiac function and tissue composition. So, that’s something we’re looking forward to digging into.
Matt Kelly: And I think on the flip side, we were surprised that we didn’t see a lot of lung involvement. So, we were limited within MR as to how we can characterize the lungs. So, we can’t look at clots, for example, or we don’t necessarily have the sensitivity to fibrosis in the lung that’s been reported. But still, we were quite surprised that it was so low given the role that breathlessness has in the disease and how prevalent that is. So, I’d say for me, those were the things that were worse than we expected and better than we’d expected.
Roger Green: Let me ask the rest of the Surfers a question. First of all, Matt, this has been a really fantastic half-hour or longer, and you guys are doing great work in listening and learning a tremendous amount, and it’s really inspiring work and obviously going to be very, very, very important to all of us what we’ve learned going forward. So, great for that and thanks for joining us.
Roger Green: Let me ask the rest of you folks a couple of questions. What impact do you think these results will have in terms of how the stakeholder community that you come from will look at the COVID and liver disease and how it does or doesn’t fit together?
Louise Campbell: It’s interesting because we’re all discussing if the liver and NAFLD is involved with COVID-19. And I think we’re getting more and more evidence that suggests that it’s involved in some respects, but we don’t know the mechanism of how it’s involved. There’s obviously the suggestion of the inflammatory effect that goes on. And in fact, I was reading earlier today the fact that that inflammatory effect could worsen NAFLD and NASH or speed up NASH. And I wondered whether or not looking at the pro scan as to whether or not that’s the sort of thing that the longitudinal aspect of this study might be able to detail a little bit more for us, whether or not we see more liver damage in some of these patients as we get further post-COVID, or whether or not we get a resolution. I just wondered what Matt thought the possibility of that was, or Steven or Donna, really.
Matt Kelly: I can take first crack and then leave it to the true experts. What I would say, if I had to guess or put my money somewhere, I would say that the inflammation we’re seeing is probably linked to this inflammatory cascade that we’re seeing in COVID patients, the same mechanism that’s likely responsible for the neurological brain fog that’s reported and is observed. But then, as you know, everyone knows very well these inflammatory cascades, once they’re triggered, it can be very tricky to predict what their course is going to be, whether they’re going to resolve by themselves, or whether they’re going to need management, or is there some sort of tipping point where the cascade has gone into freefall and will need intervention. So, I mean, that’s something the follow-up scan will hopefully shed a little bit of light onto, but more studies, more research.
Stephen Harrison: Yeah, I’ll follow on with that one. This is Stephen. So, maybe just to extend that thought a little bit and then twist it around and see what lessons can be learned. So, you scan these people on average three months after discharge from hospitalization, and the ones that were the sickest, and still the CT one was higher than those that didn’t get hospitalized, if I heard you correctly.
Stephen Harrison: Which makes me think the liver is incredibly dynamic. It heals itself very well. If you give it any time to recover, it will. Alcohol is a great example of that. So my guess is the CT-1s in hospital were probably dramatically higher, suggesting and correlating with their potential cytokine storm or a cascade of cytokines that they were experiencing that led to the reason for their hospitalization. We know that the liver, when it’s inflamed, is generating a significant amount of chemical signatures that stimulate downstream effects on other organs outside the liver, but even within the liver.
Stephen Harrison: And I’m wondering, now that I’m sitting here for the past forty-five minutes thinking about this, if the therapies that we’ve learned to treat COVID with, things that shut down the inflammatory cascade, this cytokine storm if you will, if we can use reverse engineering to target fatty liver disease with very similar modalities. I know it’s kind of a weird way to think about it. But did we just hit on something where we can actually learn from what you generated and learn what we know about COVID and apply what we’ve learned with the potential therapies for COVID, to take that back to the bench and apply that to NASH? Wouldn’t that be interesting if we could do that?
Donna Cryer: I certainly hope so. Thinking of this from rather a macro view, since it’s my job in advocacy to look for opportunities to elevate and to unify. And also my perspective as a patient who comes to this from an auto-immune perspective and who has been immunomodulated post-transplant now for twenty-six years. And so I’m always asking the question about, is it an inflammatory factor? Is the immune system involved? Either from a primary basis or just figuring out how something will apply to those whose immune systems are challenged in some way. And it has been one of the definite insights of the Global Liver Institute’s COVID-19 response program that this information is so impactful across the liver community, across the transplant community, across those with cancer, across those who are immunocompromised.
Donna Cryer: So from a patient perspective, we seek out and look for these cross-currents and applications. At least today, as we were recording, we got the information about the wonderful, delightful Harvey Alter and his colleagues being awarded the Nobel Prize for their discoveries in hepatitis C that allowed for development of a cure in such a short timeframe when you talk in the scope of medical history. Unfortunately the hepatitis C example did not do was to create an opportunity for discussion and movement on other liver diseases and opportunities in liver health. Although, I think it certainly attracted more investment and research interest. But I think NASH is different. And what we’ve been talking here today, and what Stephen pointed out, was such a perfect example of how and why NASH is different and why we’re on the precipice of positioning liver diseases plural liver health, and the centrality of the liver at an elevated position in the global health conversation.
Donna Cryer: So if all of the millions of people around the globe who have unfortunately been affected and infected with COVID are now liver patients in some form or fashion, that gives us a much larger platform and attention and spotlight to be able to get traction on so many of the issues. And whether they’re research projects or advocacy or access issues that we have been trying to advance in the liver community for years. So, I think this is just the right type of thinking about what can we learn from COVID that can be applied not only to NASH but to other liver diseases?
Donna Cryer: And what have we learned from other diseases, and what other disciplines and what other approaches, whether it’s infectious disease or immune system modulation, can be applied and pulled through? So to me, this is the most exciting type of conversation and the most meaningful type of conversation, because these are the type that we will look back hopefully on these moments ten years from now and say, “Well, this is when it really catalyzed and changed a lot of thinking and created an opportunity for us to do so much more than we had been before.”
Roger Green: So, I have a sneaking suspicion we could go on for another hour now, except we’re going to run out of schedule and our listeners are going to run out of patience. So Donna, not surprisingly, that should be the last word. And that was a great place to end today. Matt, this was just brilliant. Please come back in three months after you’ve seen the six month scan results, and let’s repeat this experience. Because I’m sure there’ll be a lot more to learn then and really wonderful things to learn. So thanks so much for being with us today. Okay. With that, let’s move towards wrapping up. The question we close with every week I think we’ll use again in some format. Two options. Number one is, what’s the thing that you heard today that surprised you the most, or what’s the most important thing that you learned today? Brave one, go first.
Louise Campbell: I’ll be the brave one on this one. And I think it’s just a summation of Matt’s and his team’s work, and what Donna and Stephen were saying there. That if we assume now that every patient that suffers from COVID and potentially is hospitalized becomes, with this study and with other mounting evidence, we identify them as liver patients, how are we really going to be able to assess these patients going forward? It does give us a better platform. And hopefully we can take it forward better than it was done in SARS, MERS and H1N1, and get that momentum. So I think it’s just the summary that all of these patients are potential liver patients, which is amazing and frightening.
Stephen Harrison: I was surprised by the cardiac findings that Matt reported. And maybe I shouldn’t be, knowing what we know about the liver and heart disease, not so much related to coronary artery disease, but what we’re learning about something called HFpEF, which essentially is stiff right ventricles that set the patient up for being at increased risk for right-sided heart failure. And so I’m really intrigued by that data. And Matt, hopefully you can shed more light into what that looks like at the six month time point and maybe a deeper dive into the data you have now relative to that stiff right ventricle. And is it validated? And is it actually linked to worse presentations of COVID in the setting of fatty liver disease? So again, fascinating conversation today. Thank you so much for coming on and sharing your insights.
Donna Cryer: My takeaway is just an excitement that we have an example that beyond the biopsy is more than just a slogan, more than just a campaign, that it is something that people are implementing and executing in real life that is having impact, having publications, changing science, potentially changing care. And so, that is the highest honor and the only legacy that an advocate can hope for. So that’s my takeaway and my excitement and joy today.
Roger Green: Hear, hear! Matt, what’s yours?
Matt Kelly: I would say my excitement is always talking to people who are so passionate about a cause, especially something as important as this. And hearing all about the time and the effort and the energy that they invest, and really push things forward and really make a difference in individuals’ lives, but also the community as a whole. So I always find it hugely rewarding talking to and meeting people who are doing that. So just spending this short time on the call with you all has kind of really given me that sort of warm glow. So, thanks very much.
Roger Green: Oh, and by the way, whatever we gave you, you gave us back at least in triplicate, possibly quite more than that. Stephen, I was thinking back to NASH-TAG, when you put up the paper about the kappa scores or coder reliability with biopsy, and the realization that one of the reasons it was so hard for anyone to figure out what was going on here is that our instruments were so blunt, biopsy being a blunt instrument. And that we’re working through a field of haze. The combination of everything we’ve talked about, the quality of the work that you folks have done, Matt, combined with the idea that COVID patients are liver patients, combined with all the dimensions at which liver and COVID seem to intersect says to me that we might be entering a period where we’ve gotten much better tools to gain clarity on what’s actually going on.
Roger Green: I suspect that will yield a ton of insights in different directions. Stephen, yours is about going back to the bench being one of them. Donna, yours about how do you really pay off beyond the biopsy being another. I can see this feels seminal on a whole bunch of levels. And we haven’t even gotten into some of the other ideas around it, like viral shedding, but this feels seminal. And it feels like, Matt, what you folks have proceeded to do is just provide so much clearer a way to look at and look into the disease that existed before. And I think it’s really exciting. And that’s what you’re hearing from all of us today. So, thank you for that. Thanks… Stephen, go ahead.
Stephen Harrison: Yeah, I was just going to say, Matt, before you go, send my best to Cat for allowing you to spend your evening with us.
Matt Kelly: I will do. I will do. I think she’s also working to, sadly. I will pass that on. Thanks.
Roger Green: Stephen, I don’t know if I shared this with you, but Matt has not listened regularly to the podcast, but Cat does.
Matt Kelly: I’ve corrected that situation over the weekend for the record, and will continue to.
Roger Green: Bravo. Now, the only thing you owe me for coming on is you owe me a picture of what social distancing Taekwondo with kids looks like. I still can’t get that one out of my head.
Donna Cryer: I think it’s like miming.
Louise Campbell: Probably more like tai chi.
Roger Green: In my head, I got Ralph Macchio and Pat Morita going, “Wax on, wax off,” but I’m sure that’s not exactly what it is.
Matt Kelly: It’s sort of enthusiastic line dancing.
Roger Green: Well Matt, thanks. All of you. This has just been brilliant. Thanks for a great episode. I think people are going to have a wonderful time listening to this. Thanks to our team, MiC Wilson, who makes this all sound so good. Buzzsprout, who puts it out there. Eric, who coordinates. Poli, who makes sure that everybody gets microphones in and out, and makes that work. Ryan Segura, who edits. And all of you folks who listen. Listen to this episode and tell all your friends. I think there’s something really special here this week. I look forward to seeing you all next week. We’ll talk more about AASLD and we’ll come up with the thing I thought we might do this week until this emerged but we have a lot of exciting topics ahead. For everybody, stay healthy, stay safe, surf on, see you on the podcast next week. Bye-bye now.
