Welcome to the 2021 postgraduate course. My name is Laurie DeLev from the University of Southern California. I'm President-elect of ASLD and the director of this year's postgraduate course entitled The Epidemic of Fatty Liver from Obesity and Alcohol. Given all the activity in the field of fatty liver, I hope you agree that the time is right for the postgraduate course to focus on one liver disorder, fatty liver disease. The course is always a group effort, and my fellow course organizer, Mary Rinella, and I would like to thank those that have contributed to the planning and organization. First, the members of the Annual Education Committee who reviewed the course in-depth and provided invaluable suggestions and feedback. Second, two members of the ASLD staff in particular, Jessica Jessop and Stephanie Grimsby, were essential to organizing the course and keeping all of us on track. It is a large amount of work, and they did the work efficiently, helpfully, and always with good cheer. We couldn't have done this without them. We would also like to thank Dr. Philippe Antoinette, a world authority on alcohol-associated liver disease, who provided the first outline of the alcohol portion of the program. Finally, I really don't have the words to adequately thank Mary Rinella, my co-director of this course. Dr. Rinella is not just an expert of NAFLD, but also has a broad knowledge of the current issues in the field, knows a wide swath of the investigators working in NAFLD and worked tirelessly with me to put this course together. If you'd like the choice of speakers and topics, the lion's share of the credit goes to Dr. Rinella. The course speakers are all experts on their topics and experienced speakers. At the time I recorded this message in mid-October, I had not yet watched the course, so I am very much looking forward to what we are about to watch. If you're watching this as it is broadcast during TLMDx, please note that after the course ends, there will be a 30-minute break followed by a live Q&A session with speakers and session chairs from 6.30-8 p.m. Eastern Standard Time. The Q&A will be divided into three separate sessions to avoid having too many people on at once. Now, forward to the course. I'm sharing the first session with Michelle Long. There are three talks in this session entitled NAFLD, Who's at Risk and Why? The first talk will be by Professor Quinton Anstey. Professor Anstey is Chair of Experimental Hepatology at New Castle University, and Deputy Dean of Research and Innovation for the Faculty of Medical Sciences. His talk is entitled, The Good, the Bad, and the Ugly, Risk Stratification in NAFLD. Our second speaker is Dr. Arun Sanyal, Chair of the Division of Gastroenterology, Hepatology, and Nutrition at Virginia Commonwealth University, and I might add, a former president of ASLD. Dr. Sanyal will be giving us the 40,000-foot perspective on where we stand, with a talk entitled, Actionable Targets of Disease Progression in NAFLD. The third speaker in this session is Dr. Miriam Voss, from the Division of Gastroenterology, Hepatology, and Nutrition in the Department of Pediatrics at Emory University, and Co-Director of the Center for Clinical and Translational Research. The title of Dr. Voss' talk is, Pediatric NAFLD Stress Conception and Before. Hello. My name is Professor Quentin Anstey from Newcastle University in the United Kingdom. I'd like to start off by thanking the organizers for the invitation to talk today about risk stratification in non-alcoholic fatty liver disease, the good, the bad, and the ugly. These are my disclosures. So the first thing when we're considering biomarkers for non-alcoholic fatty liver disease, is really to come a step back and consider what it is we want to do. And really, we need to found this in an understanding of the disease natural history. We need to break down the cases with fatty liver disease into those most likely to come to harm and those least likely to come to harm in the long-term. So in many ways, the good are those with simple steatosis, or maybe a degree of steatohepatitis without significant fibrosis. The bad, that is the increasing scarring of the liver. Either clinically significant fibrosis greater than F2 or advanced fibrosis greater than F3. And then there's the ugly. And that of course is cirrhosis, F4 disease. So when we're considering fatty liver disease, what we want to do is identify those individuals with significant, advanced disease or cirrhosis, and to use non-invasive tests where possible to achieve those goals. And the reason for that is because we've seen now in a number of longitudinal studies that all cause mortality and liver mortality really begin to take off at F2, and particularly beyond F3 fibrosis. Whereas you can see here, the risks of mortality begin to take off exponentially. Up till now, there's been a heavy reliance on liver biopsy. And we've used that as a diagnostic tool, a prognostic tool to monitor disease progress and also in clinical trials to judge whether there's been an improvement with therapy. But there are of course issues with liver biopsy in terms of tolerability, sampling error, interpretation, and so on. And so a plethora of different non-invasive biomarkers have been proposed across the full spectrum of fatty liver disease. We are going to particularly focus on those addressing fibrosis and cirrhosis, because as I've mentioned, these are the biomarkers most likely to give us information, which will guide our understanding of outcomes for our patients. And as you can see, these fall into two broad categories. There are the wet biomarkers, the blood-based biomarkers, and the dry biomarkers, the imaging biomarkers. Before we go too far into discussing those, however, I'd just like to put forward two concepts that are very important for us when we want to understand how biomarkers perform. The first is to consider the setting and the goal of the diagnostic test. So the way we use a biomarker in primary care, where there tends to be a low prevalence of advanced disease is very different to secondary or tertiary care where the prevalence of disease increases. And similarly, the goal, the thing we want to get out of it is different. In primary care, by and large, we're looking to exclude severe disease. So we want a high negative predictive value, whereas as we move into secondary and tertiary care, we're wanting to identify the individuals who need therapy or need increased surveillance for complications. And so high positive predictive value becomes increasingly useful. So one test may not fit in every environment. The second thing I want us to consider is the fact that some biomarkers have a single threshold, whereas others use two thresholds, one with a high sensitivity to rule out disease and one with a high specificity to rule in disease. And those biomarkers that use two thresholds suffer from the problem of an indeterminate zone where a clear diagnosis cannot be made. There are ways around that, and that's one of the things we're going to talk about a little bit as we go forward. So the first of the biomarkers I want us to consider, which really are the mainstay of our work in fatty liver disease, are the indirect biomarkers and in particular, the simple scores to detect advanced fibrosis, such as the FIB4 score. Now, FIB4 is based on very easily available parameters, age, ALT to AST ratio and platelet count. And one of the features here is that the high sensitivity lower cutoff has a very high negative predictive value. A level of less than 1.3 in people under the age of 65 and less than 2 in people over the age of 65 has a high negative predictive value, allowing us to effectively rule out advanced disease in these individuals. I tend not to look so much at the high specificity rule in cutoff simply because the capacity of non-invasive tests to positively diagnose is actually less good than their ability to rule out advanced disease. And what we know is that FIB4 and indeed the natal fibrosis score have also got strong evidence as prognostic biomarkers and so are able to predict which of our patients are going to come to harm in the long term. And beyond that, we can also, based on these recent data from Sweden, appreciate that we can use tests like FIB4 longitudinally to actually monitor our patients and identify those who are progressing towards trouble in the future. And so in this study, with a median interval of follow up of 2.4 years, what we can see is that a one unit increase in FIB4 carried with it a hazard ratio of just under 2, 1.81 for the presence of severe liver disease. And also individuals who transit from low risk to indeterminate to high risk have evidence of increasing hazard ratio for advanced disease. So now we have a noninvasive that is simple and cheap and available as a diagnostic, a prognostic and a monitoring tool. That's a great foundation for what we need. But of course, we want to find ways of minimizing the indeterminate zone with biomarkers. And one of the ways that we've modelled using data from clinical trials like the STELR studies was the use of serial biomarkers. FIB4 score followed here by either the ELF test or FIBRA scan. And what you can see on the right hand table is that by using a combination, one is able to effectively shrink that indeterminate zone without overtly inflating the misclassified cases beyond the sampling error that you'd see with any histological reference standard. So combinations of tests applied in a sequential fashion maintain sensitivity and specificity, reduce that indeterminate zone and maintain a stable amount of misclassification. Therefore, a very good way forward. What could we use as that second line test then? Well, there are a couple of options. Let's first of all, think about blood based biomarkers. Here in particular, direct biomarkers of collagen turnover and EC, extracellular matrix formation and degradation are particularly useful. And these, of course, potentially give information not just about the current diagnostic state, but because they speak to those fluxes of degradation and formation. They also give us information potentially, which allows us to monitor disease as we go forward. One of the biomarkers that's been around for a long time here is the ELF test. I'm showing you data here from a meta-analysis conducted by the Litmus Consortium of just under four and a half thousand individuals. The ELF test is composed of three components, P3NP, hyaluronic acid and TMP1. There have been numerous cutoffs proposed for the ELF test over the years, as you can see in the gray table at the top of the slide here, some of which have higher sensitivity, others which have higher specificity. In this meta-analysis, two new cutoffs were developed, optimized for either a 90% sensitivity at a threshold of 8.1 or a 90% specificity at a threshold of 10.18. And these perform relatively well in those senses. But, and this is crucial, if you look at the table in the bottom right hand corner here, you can see that prevalences of advanced fibrosis, such as we may see in primary or secondary care, 5%, 10%, the negative predictive value is very strong, but the positive predictive value of these tests remains modest at most, going up to no more than about 40%. So again, a useful tool for ruling out advanced disease and also a tool which has prognostic value, particularly in patients who've already had a liver biopsy. And that's an important point to also mention. Now, there are already data about using the ELF test in a sequential care pathway. This is some data from London, which compared essentially standard care, no prior triage, with a combination of Fib4 score followed by ELF test. And implementation of this was able to change from the referral patterns so that it went from only 5% of individuals with advanced fibrosis going up to 30% of individuals. However, just a note here, 87% of the cases excluded were actually excluded by the Fib4 score rather than the ELF test. So this trial is really demonstrating the value of any care pathway rather than the chaos of no care pathway. And that's something else which is worth noting. We could also consider using imaging biomarkers as a second line test. And there are a couple of these I want to highlight. The first one, VCT or FibroScan, which is an ultrasound based tool to measure the stiffness of the liver. And there have been a number of very nice studies here. I'm showing you data from two of those, one from the United States, one from the United Kingdom, both of which demonstrate very similar areas under the receiver-operated curve in terms of performance for advanced fibrosis. And what we can say with FibroScan is that a liver stiffness of less than about 8 kilopascals, provided it is a reliable reading, has a very high negative predictive value. Once again, able to exclude advanced fibrosis. MR elastography is a more advanced technique that is particularly getting traction in the clinical trial space, but is also becoming more freely available for use in clinical settings. And once again, a very effective tool for identifying fibrosis and probably at trials also more sensitive for judging changes in fibrosis, which is very useful. However, when we compare VCT and MR elastography in terms of those real world questions of differentiating stages F0 and 1 versus 2 to 4 or stages 0 to 2 versus 3 to 4, the performance of VCT and MR elastography is actually fairly similar in that day-to-day practice. And so whichever one of those one can access is probably sufficient. There's been a move recently to try to improve imaging biomarkers further by combining them with blood tests. We've seen the FAST score combining FibroScan with AST levels and a number of other indices to identify patients for clinical trials. We've seen a further study called MIRNA, which is an MR elastography and Fib4 combination, which appears to outperform the FAST score. And here I'm showing you data which is actually being presented at the liver meeting from the Litmus European Naffled Registry cohort, looking at two new scores, Agile 3 and Agile 4, which again are based on FibroScan. And what's happened with these is that the sensitivity and specificity has been adjusted specifically to optimize the positive predictive value of these tests, either for greater than F3 disease or cirrhosis in the case of Agile 4. Now, the interesting thing about this is that both liver stiffness and Agile outperform Fib4, as you can see very clearly in the AROC curves here. But it's also the case that the Agile scores do appear to outperform liver stiffness measurements, certainly in terms of their positive predictive value. However, there is one caveat here. Unlike a single threshold, as we might use for the simple interpretation of FibroScan liver stiffness, Agile 3 and Agile 4 are again using those two separate thresholds of high sensitivity and high specificity. And so the greater positive predictive value is generated, but at the cost of developing an indeterminate zone of up to 15% of individuals who cannot be classified using either the Agile 3 or the Agile 4 scores. So a very useful tool, but we need to consider about what penalty we're prepared to pay for that increased positive predictive value. What do I do in my practice right now? Well, I like to keep it very simple. I use the Fib4 score. I use a single threshold with the Fib4 score. Greater than 1.3 if an individual is less than 65, greater than 2 if an individual is over 65 years of age. If the people fall below those thresholds, it's lifestyle advice, discharge back to primary care. Those above the threshold, I move on to a second line test. I tend to use FibroScan. One could use ELF test and a threshold of about 8 for either of those is very tractable. Again, allowing you to rule out individuals and focus in on the ones most likely to have advanced disease. And I reserve liver biopsy for those minority of cases where I believe it is going to add extra value. Having leveraged that high negative predictive value of two serial rounds of testing to allow me to identify those individuals. In the last few minutes, I'd like to talk a little bit about some of the exciting new developments and what's on the horizon in terms of fatty liver disease, what we may well see in our practice in the next few years. I think we're moving now away from those simple tests. They are the foundation for now, but the future is likely to be a knowledge driven paradigm for fatty liver disease management. The first part of that is about a priori ascertainment and mitigation of metabolic risk. So it's a form of primary risk stratification before disease maybe has even developed. And here we're really talking about the use of polygenic risk scores generated using PMPLA3, TM6, GCKR, MBOT7 and HSD17B13 as examples of some of the modifiers that have been identified. And use of these scores is able to identify individuals with up to a 12-fold increased risk of advanced fibrosis or hepatocellular carcinoma. So promising data there. Then the second part of that knowledge driven paradigm will be about a personalized medicine approach using omics technologies to devise new biomarkers that are better for diagnosis, staging and monitoring disease. And in this respect, we're already seeing this beginning to come through. Transcriptomic profiles of liver tissue across the fatty liver disease spectrum have allowed us to identify a 25 gene signature, characterizing and differentiating different degrees of severity of underlying fibrosis. And one has been able to move beyond those to then look in the circulation to find specific circulating biomarkers derived from the liver that guide us to disease severity. Here's one example of that, which is GDF15. Another is AKR1B10. And there's been further work which has been done by our group collaboratively with colleagues working in Japan, looking at the identification of variants in thrombospondin 2. Again, taking us from transcriptomics through to protein signatures. Another approach is to go straight in with proteomics, and a great example of this is work done using the SOMA scan platform, where a number of scores have been identified that identify severity of steatohepatitis and stage of fibrosis. Now, further data on these was presented at the European International Liver Congress in the spring of this year. And there are also a number of presentations here at ASLD demonstrating how well these particular platforms can perform. The final novel biomarker is one addressing the activity of proteases within liver tissue. Now, this one is highly novel and as yet at a very early stage of development. But I just include it here because it is such an interesting and exciting way forward. So proteases are fundamental to biological processes, but actually are very hard to access because they tend to be tissue bound rather than released into the blood. So one needs to find ways of getting to them to assess their activity. And one tool is essentially to inject patients, as you can see on the right hand side here, with particles somewhat like a hedgehog covered in protein spikes that then travel to the liver where they are cleaved off by active proteases. Now, each of these spikes carries a specific protein mass barcode to it. And so when it is released in the circulation, that spike is then passed to the kidneys and passed out in the urine. And so by analysing the urine by mass spectroscopy, one is able to actually then identify which proteases are active in the liver and which of those spikes have been cleaved off. Now, in collaboration between Newcastle and the Mass General Hospital cohorts, we've been able to provide transcriptomic data to help guide which of the proteins should be put on those spikes. And this has gone forward into further development by Glymphs Biosciences and a paper is just coming out now in Science Translational Medicine demonstrating in animal models how these mass protease barcodes can be used both to differentiate fatty liver disease from healthy and also potentially to monitor sensitivity to change in a treatment response paradigm. Very early for these sort of data, but a fascinating way forward. So, to conclude, fatty liver disease is highly prevalent, largely asymptomatic, and characterized by substantial interpatient variation in disease severity and outcomes. There are a range of biomarkers, indirect and direct serum biomarkers, and imaging biomarkers. At present, combinations such as the FIB4 score, followed by a tool like FibroScan or ELF, appear to be some of the most stable, but this field is evolving incredibly rapidly, and new biomarkers are being devised all the time. We have an urgent need for better biomarkers, but we also have a pressing need to validate and assess the performance of biomarkers objectively and robustly, and this is something that international cohorts like the FNIH Nimble Consortium and the EU-IMI2 Litmus Consortium are currently working on. It leaves me, then, to thank you very much indeed for your attention, and to hope you enjoy the rest of the conference. Hi, my name is Arun Sanyal. It's a pleasure for me to talk to you today about actionable targets of disease progression in NAFLD. Here are my disclosures. Now, non-alcoholic fatty liver disease is a heterogeneous disorder. It is heterogeneous clinically in terms of its progression, with some people following a relatively benign course, while others have a more progressive course, and yet others who actually can spontaneously improve. This cartoon sets the stage for our conversation today, and describes our current thinking about how this heterogeneous disorder evolves. Starting at the top, in the right genetic setting, when one develops increasing adiposity, it triggers accumulation of fat in the liver. This produces changes that lead to disease perpetuating mechanisms being activated, while at the same time, tissue healing mechanisms come into play. It is this balance between injury and tissue healing that determines the phenotype and its progression, so that when tissue injury patterns are the predominant pattern, disease progresses, whereas if tissue healing progresses, then that leads to restoration of normal phenotype and function. So, let's start at the top and talk about genetics. Multiple genes have been related to development and progression of NASH. Of these, the PNPLA3 is the most important, and has been most validated. Now, it is interesting that virtually all of these genes are linked to lipid trafficking proteins, and for a long time, there has been a paradigm that steatosis is not relevant in NASH. Now, that's probably incorrect, and over here, what is shown is really the changes in steatosis versus fibrosis, and the impact of various genetic markers. It does show that as goes the steatosis, so goes the fibrosis. HSD17B13 is somewhat unusual as a variant, because it is the variant which is protective and has retinal dehydrogenase activity and may protect from NASH. Now, we're learning a lot about PNPLA3 and particularly about HSD17B13 in terms of exactly what it does that leads to disease progression or protection from disease progression. So, in the setting of PNPLA3, as shown in this cartoon, you can see that there is an increase in oxidative stress, diacylglycerol, ceramide formation, which triggers immune activation on one hand, and inflammatory pathways on the other that converge on a TGF-beta-mediated increase in fibrogenic signaling. And this is shown on the right in terms of actual gene expression studies when you overexpress PNPLA3 and can actually accelerate the disease phenotype in appropriate animal models. So, having talked about the genetic background and how gene expression levels are now being targeted for therapeutic gain, and early phase trials are already underway for PNPLA3 and HSD17B13 to either knock down the PNPLA3 expression level or knock down the expression of the wild-type HSD17, which is the more harmful variant. Let's move on to sort of more common and garden-variety issues, which is the disease initiator, which is that increased adiposity. Now, when you shift your diet to a more obesogenic diet, and as fatty liver develops and progresses, there are shifts in the microbiome. There are already changes in the microbial structure that have led to identification of both microbiota and microbial metabolites, particularly in those with advanced fibrosis, that can help identify the subpopulation. In addition, in serotic populations, the risk of mortality, ACLF, and ICU transfer for patients admitted to the hospital can also be identified via the microbial composition. But in my view, this is still early days. We still have a lot to learn about how the microbiome actually modulates the development and progression of disease, and the few handful of trials that are currently ongoing include probiotics, et cetera. But in my view, there's a relatively crude approach to a very complex problem, and much more powerful approaches using FADGE-based therapeutics, et cetera, are yet to come. Now, another thing that happens when you start gaining weight, which is at the root cause of development of fatty liver disease, is that the duodenal mucosa hypertrophies in response to a greater load of easily digestible foods, whereas the mucosa in the colon tends to thin out with corresponding changes in gut permeability. In addition to the hypertrophy in the proximal gut, the number of neuroendocrine cells, as shown on this slide, also tend to become more dense in those who are on high obesogenic diets, and signaling from these is considered to be an important part of the progression to obesity and resetting of the metabolic machinery of the body. Much of this work came from the bariatric literature, where proximal gastric bypass leads to rapid improvement in metabolic state and has led to now approaches to resurface the duodenal mucosa for metabolic gain. And as shown on the right, you can see significant improvements in hepatic steatosis when the duodenal mucosa is resurfaced compared to sham-operated. Now, some of the other endocrine signals feed into the brain, and a number of mediators from the gut to the brain have been identified that have important effects on food intake. Of these, GLP-1 is probably the most important one from an actionable point of view and is already in phase 3 trials for the treatment of NASH. GLP-1s have major effects on the cardiovascular system, as well as on the kidneys, and now we're learning they can improve steatohepatitis as well. So, phase 3 trials are already underway to see if the improvement in GLP-1 signaling through exogenous drugs can not only reduce weight, improve cardiovascular and renal outcomes, but also improve steatohepatitis and progression to cirrhosis. Behavior and the brain are closely interlinked as major drivers of the development of diet-induced obesity because behavior is what drives what we eat, how much we eat, when we eat, and so on and so forth. But what we eat is modulated and that behavior is modulated by mental health disorders such as depression and anxiety, social factors such as isolation, food addictions, et cetera, all of which impact our eating behavior. Now, when you overload the gut with nutrients and you become obese, both the gut and the systemic milieu, as well as the liver disease and the liver, send a variety of signals back to the brain that can further compound this effect. So, this is an important area that underpins the behavioral aspect of not only NASH, but really its underlying obesity and other related disorders such as type 2 diabetes, cardiovascular disease, et cetera. And in my view, many of these are providing actionable targets and should be a focus of future research. An important area that has emerged is a broad dysfunction of the pituitary axis. There are early data from the HIV literature indicating that visceral adipose tissue can be increased and there's a relative decrease in growth hormone, releasing hormone, which is from the pituitary gland. And using GHRH analogs, one can reduce visceral adipose tissue as well as liver fat. Its role in garden variety NASH has to be established. Similarly, hypogonadism is extremely common, as shown in this slide, where over 60% of patients with fatty liver and histologically shown biopsy proven steatohepatitis, these are males, actually had hypogonadism. And when you correct this hypogonadism, it leads to a substantial improvement in liver fat. Histologic data will be coming soon, we hope. Now, the net consequence of all of this is modulation of the metabolic flexibility for whole body energy homeostasis. So what is metabolic flexibility? Metabolic flexibility is a state where under low infested conditions, you have low insulin levels. And under these conditions, you release free fatty acids from adipose tissue, muscles utilize them, and the liver during this time makes a little bit of glucose to feed the brain. Now, when you eat, what happens is you get a surge of insulin, which shuts off lipolysis, and the insulin drives glucose into the muscles and the muscles switch back and forth between fat and glucose. And that's what's called metabolic flexibility. Now, when you become metabolically inflexible, you have high levels of insulin all day long. The adipose tissue is resistant to the lipolysis suppressive actions of insulin, and you continuously leak free fatty acids. But the muscles are unable to oxidize these. And in the postprandial state, because of insulin resistance, the muscles cannot clear glucose. This excess free fatty acid and glucose, which would normally have been metabolized in muscle, now are available to the liver and contribute to the lipotoxic and glucotoxic loads and feed the triglyceride overload, which is the hallmark of NASH. So, some key takeaways are that gut-derived signals to brain from microbiome neuroendocrine cells modulate eating behavior, and multiple brain functions with respect to metabolic homeostasis are altered in the obese insulin-resistant state, and increased adiposity and altered neuroendocrine regulation of metabolism lead to metabolic inflexibility that drive fatty acids and sugar to the liver. Now, let's now move on to the liver related factors themselves. We've already talked about the impact of dietary fat, and now we've talked also about delivery of excess dietary sugars, which through a process of de novo lipogenesis under the control of SRE-BP1 can lead to the fatty acid pool in the liver that drives triglyceride levels. These can be modulated by a variety of factors, and these are shown in these boxes on this slide. Of these, all of these virtually are in clinical trials in various phases and represent actionable trials to reduce the metabolic overload of the liver. Now, a consequence of this, of metabolic overload, is steatosis, and as I said earlier, the paradigm that steatosis is irrelevant to the disease is probably incorrect because even neutral fat not only has been linked to the changes in PNPLA3, etc., but also to the activation of downstream injury processes. Now, in addition to the triglyceride and fatty acids, there are other lipids such as eicosanoids, free cholesterol, ceramides, phospholipids that are all altered in NASH, and when you overload the system with these metabolic products, it stresses the cell because of the need to address this metabolically. Now, if you can adapt, you can remain relatively healthy, but a maladaptive cellular response leads to disease and to organellar injury, which together lead to activation of inflammasomes and the innate immune system. Overall, this concept is known as metabolic inflammation. Now, metabolic inflammation is not unique to NASH. This is part of our normal aging process where inflammation sits in the middle of metabolism, proteostasis, macromolecular damage, cell stress, epigenetics, and stem cell regeneration, and creates what we call a bow-type shaped inflammatory response where, on one hand, you have inputs, which is called FANIN, shown on the right, from non-cells, that is PAMPs, quasi-cells from nutrients and microbiota, and cells from damage-associated signals or from the tissues that activate a limited number of sensors, which then lead to a FANOUT signal, leading to a larger inflammatory response. And this is accelerated in the context of NASH. So normally, the process that you see in aging is amplified in NASH. So in many ways, you can consider NASH as a prototypic disorder of accelerated aging via the process of inflammaging, which then provides multiple potential targets for intervention. At a transcriptomic level, as you look at progressive disease, you can see many of the different pathways that are involved at the cellular level and how they interact with each other through network analyses. And these are provided not only the traditional targets, but a number of downstream targets related to the immune inflammatory response that may be very relevant for fibrosis progression and need to be studied further. And truly, taking this one step further, within people who have similar fibrosis stages, based on the transcriptome, early data suggests that there may be clusters of individuals who have unique patterns of gene expression, leading to the idea that if you identify the key drivers in each of these clusters, you might be able to approach this disease in a more precision manner, allowing targeted therapeutics to be brought on board, which would be likely to be more effective. New data from single-cell RNA-seq is also providing novel insights into NASH. So these are some early data showing different metabolites and how they are interlinked to various cells and cell functions, and particularly connecting the hepatic stellate cells, HSC, with a number of immune inflammatory cells. And these have established particularly a role for TREM2-positive cells, where this TREM2 expression has been linked not only in the context of the disease itself, but to AST, ALT, steatosis grade, ballooning grade, and even fibrosis grade. So we are now getting a much more granular understanding of the immune inflammatory response, secondary to the metabolic overload and injury that occurs more upstream. And of course, the flip side of this is the regenerative response. And we know that dysregulated tissue repair and activation of development pathways underlie disease progression in NASH, where lipotoxicity with tissue injury from ER stress and mitochondrial dysfunction, along with steatosis, lead to activation of NOTCH, YAPTAS, et cetera, on one hand, and on the other hand, trigger cell death. Now, downstream of YAPTAS signaling are additional pathways that can converge on TGF-beta as well as signaling from injury patterns through hedgehog ligands, which come from balloon cells. Now, these, of course, can lead to quiescent hepatic stellate cells being activated, leading to scar tissue and further inflammation. On the other hand, at a hepatocyte level, compensatory proliferation leads to new hepatocytes, but with the risk of genomic instability and development of hepatocellular carcinoma, particularly in a tumorigenic microenvironment. So the key takeaways are that the core paradigm, which has been available now present for now almost a decade of metabolic stress leading to cell injury and apoptosis, driving an inflammatory response, and eventually fibrogenic remodeling towards to cirrhosis, still remains intact, but we have much more granular understanding of the types of metabolic stressors, the extracellular extrahepatic targets that are particularly relevant, growing emphasis on the gut and on the brain as important drivers of the extrahepatic milieu, in addition to adipose tissue and muscle, and then the cellular pathways that lead to progressive disease within the liver and leads to fibrosis. So I would like to summarize by saying NASH is a disease of nutritionally driven acceleration of inflammatory aging processes. NASH progression may be halted by attacking upstream drivers, producing a fibrosis break, or a combination of these approaches. Novel insights on heterogeneity of disease states may allow precision approaches to slow down disease progression. I thank you for your attention. I'm delighted to present in this exciting post-grad course on NAFLD, and thank you to Drs. Ronella and Daliv for inviting me to participate. Here's my disclosure slide. There are a few studies that follow the trajectory of NAFLD from conception to adulthood, and in fact there's no studies that follow all the way to middle adulthood and later adulthood. However, we can start to piece together NAFLD across the lifespan using data that exists. This figure depicts this trajectory from in utero influences on the left hand side and maternal fatty liver to early childhood insulin resistance and hypertriglyceridemia. Most children are diagnosed with NAFLD during the school age years, peaking between age 10 and 12. It's from here that they can go on to develop the associated conditions of type 2 diabetes, cardiovascular disease, and later on potentially the adult outcomes of cardiovascular disease, cirrhosis, hepatocellular carcinoma, and liver failure and need for liver transplant. So first, let's address maternal influences on pediatric NAFLD. There is a growing interest in the maternal drivers of pediatric NAFLD, and some of the most compelling animal model data comes from non-human primate studies. In the study shown here, pregnant monkeys were fed either a control healthy diet or a high fat diet. Importantly, the investigators divided the high fat diet mothers into a high fat insulin resistant group and a high fat insulin sensitive group. The histologic outcomes of the offspring is demonstrated in the slide below. The offspring were also divided into two groups. Some of the infants were given control diets, which were healthy, and some were provided continued on the high fat unhealthy diet. What's shown here is the histology from the here from the mothers who are on the control diet and the infants on the control diet, so healthy healthy. Here is high fat insulin sensitive mothers and control offspring healthy diet. And then lastly, here's the high fat with insulin resistant mother conditions and a control diet. Critically, these infants were provided what's considered to be a healthy diet, but yet they had steatosis present in much greater levels in the liver, regardless and despite of that healthy postnatal diet. This study was an important step in understanding the critical nature of the maternal condition and driving pediatric NAFLD, and that potentially irreversible changes may be occurring. High plasma triglycerides and cholesterol are high normally in a physiological normal pregnancy. However, circulating triglyceride levels are even higher in mothers with gestational diabetes mellitus or obesity. Higher free fatty acids are found in umbilical cords from mothers with gestational diabetes, and there's evidence for crossing of the placenta for lipids. These metabolites and other cytokines, hormones, hyperinsulinemia, obesity-induced hypoxia likely alter the fetal liver conditions. Other studies from this same group, as the references are shown there, have also suggested that the maternal obesity condition may transmit an altered microbiome to the offspring, shown here at the bottom in yellow. See, let me do my laser, here we go. Offspring gut dysbiosis, offspring gut permeability, and then gut-derived factors may go together with the altered metabolite milieu to generate altered hepatic macrophage programming in the fetal and neonatal liver. This altered programming may be what goes on to react to the postnatal or childhood high-fat and high-sugar diet, and then resulting in increased inflammation and steatosis postnatally, and the disease, NAFLD and NASH. This figure summarizes a large number of compelling animal studies, and I refer you to the two references to read more about them. So what about maternal obesity and NAFLD in children? So pre-pregnancy obesity is going up significantly in the United States. We have data shown here from 2016 to 2019, and the rate of preconception obesity went up in all age groups except for the teenagers, which is shown here in the gray line, the under 20. I guess the teenagers are relatively preserved, although they're going up too. This group, which is really the majority of pregnancies, the age 30 to mid-40s, represented 29% of that group was obese at the time of conception. So does it matter if the mother is obese? Well, this is data from a population-based case control study using the EXPRESSO cohort from Sweden. NAFLD was able to be determined in offspring using medical records. So they had medical records where they could look at biopsy-proven NAFLD from 1998 to 2016 and match that to maternal records of body mass index in early pregnancy. The odds ratio for any NAFLD in offspring was three-fold greater if the mother was obese. And importantly, the odds ratio for fibrotic NAFLD was also increased, 3.67-fold increased when the mother was obese. There are a number of studies looking at maternal and paternal factors, and the strongest risk factor is maternal pre-pregnancy BMI. So if the mother's obese, that's the strongest risk factor for childhood NAFLD in both boys and girls. Now, not to let dads completely off the hook, there is some data coming out of the Australian RAIN cohort that a higher paternal BMI was strongly associated with NAFLD in male adolescents. But interestingly, not in female adolescents. So maternal obesity, is it a multifactorial effect? Is it primarily the shared microbiome? Well, while much progress is being made on the mechanisms of the maternal influences, it is likely, I think, to be a multifactorial effect. We know that NAFLD is highly genetic, and so there would be shared genes between the mother and child, obviously. And there's a shared set of environmental exposures and shared postnatal diet in most cases. So we really need more studies of the modifiable factors, particularly metabolites that cross the placenta or effects that could be modified by change in the environment or through education of the parents planning pregnancies. What happens to women known to have NAFLD? So not just obesity, but also NAFLD. Well, these mothers are already at baseline six-fold more likely to have type 2 diabetes. They're three-fold more likely to have gestational diabetes develop during the pregnancy and to have preeclampsia develop. The infant is two times more likely to be premature and large for gestational age. Both factors are associated with childhood obesity, but we don't have data on whether maternal NAFLD is programming or associated with childhood obesity. And NAFLD. What happens after the infants are born with steatosis? We really don't know. That's because there are no studies currently following infants and measuring steatosis in the liver and following them longitudinally that I'm aware of. So what about other outcomes? It's not just NAFLD that happens after an obese pregnancy. We know that postnatally, the offspring are more likely to have obesity, increased adiposity, glucose intolerance, insulin resistance, NAFLD as we've talked about, and then cardiovascular outcomes as well. Let's talk about the impact of NAFLD in children. Pediatric NAFLD is common across the world. This map shows countries color-coded by obesity prevalence, and I've superimposed the published rates of NAFLD in children upon these regions. On the left-hand side, in the North America region, six and a half, estimated six and a half percent of children have NAFLD. Europe, five and a half percent. Asia, 5.9%. You can see that the prevalences are very similar across the world. This is a recent study published in a multicenter fashion from Germany, Austria, and Switzerland in almost 5,000 overweight or obese children. They had liver enzymes collected from medical records and showed that 49% or half of them had elevated ALT. 12% of the children had prediabetes, and only one and a half percent had type two diabetes mellitus. Elevated ALT, or presumed NAFLD, did increase the risk of prediabetes and type two diabetes, but not as much as I expected, only one and a half fold. Diabetes and prediabetes are closely associated with pediatric NAFLD. This is a chart summarizing many of the studies that have evaluated diabetes and prediabetes in children with NAFLD. The prevalence of type two diabetes ranges, as you can see here, from two to 14%, and prediabetes from 10 to 23% of children with NAFLD. These numbers are strikingly higher than the figures of prediabetes and diabetes in obese children without NAFLD, suggesting that there's an interconnection between these two conditions. Children are insulin resistant during puberty as part of the normal physiologic changes needed for growing into the adult phenotype. I really like this figure because it shows longitudinal data. So this is HOMA-IR. Excuse me, I'm gonna get the laser figure out here. This is HOMA-IR on the Y-axis here. And here we have data in five-year-olds who are then followed prospectively to age 14. The boys are shown in the solid line and the girls in the dotted line here. And you can see that over time, once puberty starts around here, HOMA-IR rises in both groups. Now, this is a physiologically normal rise in insulin resistance, and insulin resistance should peak at Tanner stage three. So it makes sense that the girls go up before the boys since at age 14, most boys would still be in the middle of puberty. These figures here demonstrate that this normal physiologic event can become highly problematic when children enter puberty, already insulin resistant and obese. Figure A on the left-hand side shows lean children across the five Tanner stages. And again, you can see this rise to Tanner stage three and then a decline again. And this was not statistically significantly different at the end compared to the beginning of puberty. So their insulin resistance normalized after the end of puberty. This was not the case at all in the obese group. Here, their insulin resistance starts higher, it peaks much higher, and then it declines over Tanner stage four and five, but never returns to their own baseline. This is highly problematic because it is this then that leads to the next stages of pre-diabetes and type two diabetes. Similar to adults, NAFLD pathophysiology in children is tightly connected to insulin resistance. However, there are several important differences. One is this physiologic insulin resistance of puberty, making it critical to understand where an individual child is in the natural progression of puberty and insulin resistance. The other is that children are more likely to be very high consumers of dietary sugars. This figure here shows the cycle of high dietary sugar intake inducing increased insulin secretion on the left-hand side. And in the setting of high dietary sugar intake, insulin secretion is even higher and it drives de novo lipogenesis, even though insulin action may be impaired or resistant at the glucose pathways. This rise in de novo lipogenesis then leads to increased lipotoxicity and inflammation and other factors that then feed forward the insulin resistance, which then high insulin drives appetite and you get more intake of dietary sugar. And so this becomes a feed forward cycle of pathophysiology that children are engaged in. So NAFLD in childhood is strongly associated with insulin resistance, dyslipidemia, glycemia, hyperglycemia, and likely leads to early cardiovascular disease. There is evidence for increased risk of diabetes as I showed you in that chart, but there are a few studies that track from childhood into adulthood. So we really don't know what the risk is by developing NAFLD at such a young age. There is a rising rate of liver transplant for NAFLD in young adults. So this does suggest negative outcomes for some childhood onset of NAFLD. One of the rays of hope I think is that identification of NAFLD early may bring increased opportunity for treatment and avoiding these consequences. So going back to this cycle, is it possible to break this cycle? I'm gonna tell you about two different studies. In this first one, it's a crossover study of 15 adolescents and two types of breakfasts were compared. Both were low glycemic indexed foods, but one was also low insulinogenic. So it was called the low insulin index diet. In the figure, the breakfast is here and they fed the adolescents this breakfast. It was a crossover study. So they had got both kinds of diets and then they measured in the postprandial time and the low insulin index diet as shown here in the dark gray bars. You can see the three different time points postprandially. The low insulin index diet did have lower insulin response in the adolescents as shown here. And one of the exciting components was that they also showed a lower hunger index. They didn't have any other outcomes like longer term outcomes to see if it would have effect on insulin resistance or weight or even to never lipogenesis or fatty liver. But I think this would be very interesting because if we decrease the insulin stimulation, that could decrease all of these factors downstream. Another study that I think is highly relevant is this one. We conducted a randomized control trial to test if dramatically reducing dietary sugars in children would improve NAFLD. This was an eight week diet provision study in which children were randomized to either continue their usual diet or to consume foods with no or almost no free and added sugars. The data for each participant is shown here in this figure. And there was an adjusted mean difference at the end of eight weeks in hepatic fat by MRI PDFF of minus 6%. So in other words, the low sugar diet group decreased their hepatic fat, which you can see on the graph here. These are individual level data. So their baseline measurements are along the line and then the circle represents their eight week measurement. And there's a big difference between the left hand side, which is the low sugar diet and the right hand side, which is the usual diet. In this study, we also measured de novo lipogenesis using stable isotope tracers. And we found a 10% decrease in the rate of de novo lipogenesis in the low sugar diet group. And then also a 20% decrease in fasting insulin. So those are two examples of how we could target NAFLD in children using diet. What about maternal treatment to prevent childhood NAFLD? Well, there certainly is a growing understanding of the importance. This is a critical area that needs more research. Maternal NAFLD is very understudied and it's really unclear how to treat or approach this for the offspring. Maternal obesity has more work on it and more studies looking at insulin resistance. But there's really not a clear pathway forward yet. Weight loss prior to pregnancy is likely to help and there's several smaller studies on this. But the literature is a bit mixed on how to go about this and how much and when, how long before the pregnancy and all these kinds of factors. A healthy diet is already recommended for pregnancies, but it could be better tailored in the case of an insulin resistant mom or gestational diabetes and type 2 diabetes. Both gestational diabetes and type 2 diabetes often lead to counseling and treatment for the mother and those are probably very helpful. In one study, they looked at non-gestational mothers who did not meet the definition of gestational diabetes, but they were obese and insulin resistant. And those mothers frequently develop hyperglycemia even though they didn't meet the criteria for gestational diabetes. And the offspring at age four years had greater weight gain, so more likely to become obese and go on to develop NAFLD. So I do think that maternal conditions is an important area of opportunity for the future. So key takeaways. Maternal obesity is the biggest risk factor for pediatric NAFLD. Treatment of the mother of the hyperglycemia would be protective for childhood obesity and needs to be tested regarding future pediatric NAFLD in those same children. Maternal NAFLD, we know, worsens outcomes for both the mother and the child, but it's not clear at this point how to go about treating them and protecting them from those increased risks. Children with NAFLD frequently develop prediabetes and diabetes. This is very concerning. The cycle of insulin resistance in children with NAFLD does respond to a low sugar diet, but we used a diet provision model and it's very hard for families to accomplish this type of a very low sugar diet on their own. Longer term and more effective, easier to implement treatments are really needed, as well as maternal studies to try to decrease the onset of pediatric NAFLD. Thank you. Welcome to the second session of the postgraduate course. This session will focus on drivers of disease progression in NAFLD and show how to optimize their care. Dr. Rohit Lumba from UCSD will discuss the role of the microbiome in NAFLD to assess the extent of disease and how it might be leveraged therapeutically in the future. Dr. Ken Koozie will discuss the important role that endocrine mediators play in the context of NAFLD and how one might incorporate these concepts into clinical care. Then Dr. Kathleen Corey will discuss the complex relationship between NAFLD and cardiovascular disease and to what extent NAFLD might contribute independently to cardiovascular disease and its progression. Next, hepatocellular carcinoma in the context of non-serotic NASH has received a significant amount of attention over the past several years. And Dr. Helen Reeves from Newcastle University will contextualize this association and inform our approach into a clinical level. Lastly, despite the lack of FDA approved agents for NASH, there are numerous drugs approved for its associated comorbidities, some of which also may have benefit in fatty liver disease. Dr. Manal Abdel-Malik will discuss the role of existing therapies and what we might expect from them in the context of NASH. Thank you, and I hope you enjoy the session. I would like to thank the organizers for giving me this opportunity to present at such an esteemed forum. Really honored to be here. These are my disclosures. Non-alcoholic fatty liver disease is a complex metabolic disease with a clear gene environmental interaction that's happening. Based upon studies by several investigators who've been working in the field for many years, we have discovered that there are many genes that are associated with increased risk for developing naphrodi. And we know there are about 23,000 genes in humans. We also know that the number of microbial genes that reside within our body is significantly or exponentially larger than the number of genes that we have. And there are to the order of about 3.5 million genes. Now, naphrodi is affected by the diet. Of course, how the genes affect the trait is dependent on the body mass index, as well as the type of food we eat and potentially on alcohol use as well. What we eat really determines our microbiome and vice versa. And the extraction of energy from what we eat is also dependent on microbiome. So we're learning a lot about microbiome, especially in the last decade or so. Several seminal studies have been done linking microbiome to disease state within naphrodi and potentially associated with pathogenesis or fibrosis progression in naphrodi. And we'll discuss some of that today. What are the links that liver disease is inherently linked to microbiome? As hepatologists, we know that patients with cirrhosis have increased risk of bacteremia, increased intestinal permeability and increased LPS levels. Patients with cirrhosis have bacterial overgrowth in the small intestine. Several studies have shown that infections and sepsis are the leading cause of mortality and liver failure. Selective decontamination with antibiotics is beneficial for patients with decompensated cirrhosis. We know that we use empiric ceftraxone in patients presented with suspected variceal bleeding. Spontaneous bacterial peritonitis, clinical evidence of bacterial translocation leading to increased mortality in cirrhosis. And then fluoroquinolones for treatment of SBP. And lastly, rifaximin for treatment of hepatic encephalopathy. All these point towards a key role that the gut microbiome and its interaction with the host is playing in determined clinical outcomes in patients with cirrhosis and liver disease. Now, the idea is this can happen in patients with decompensated nash cirrhosis, but could this be playing a role earlier in the disease? And there are certain clues to it. How might gut microbiota influence NAFLD or liver disease progression? There's a possibility that change in caloric yield of diet and modulation of fat storage is dependent on microbiome. It may regulate gut permeability to release bacterial products coming to the portal circulation. Modulation of choline metabolism is well-known. Production of endogenous ethanol, regulation of bile acids, and regulation of lipid and energy metabolism. Now, are there any clues to gut liver access microbiome and microbial metabolites in NAFLD? And there are some. There's one seminal study I would call where they looked at pediatric patients with pediatric nash and compared them to normal or obese children. What they found was that the plasma ethanol levels were significantly increased in those with nash and were not in those with obesity or normal controls, suggesting that it's possible that the gut microbiota in these children with nash may be producing this ethanol leading to direct damage in the liver and fibrosis progression. This is a paper that was written a few years ago by Jerome Bossier and Manal Abdel-Malik and Ana Medill looking at patients with biopsy for NAFLD with 16S metagenomic assessment was not done in this study, but with 16S they were able to see which patients with nash and stage two fibrosis, how they were different than those with earlier stages of disease. And what they noticed was decrease in the abundance of bacteroides and rivitella and increasing the abundance of ruminococcus in those with NASH stage 2 fibrosis or higher. This is one paper where we've looked at the gut microbial metabolites. This particular study was conducted in Spain by Howells and colleagues, published in Nature Medicine, where they compared women who underwent bariatric surgery with or without naphrobin. And what they noted was that aromatic amino acid and branching amino acids were increased in patients with naphrobi. In particular, an aromatic amino acid, phenylacetic acid, was shown to induce hepatic steatosis both in vivo in mice, as well as in vitro in human hepatocytes, linking this to naphrobi. Now if this metabolite is linked to naphrobi, is it also somehow associated with fibrosis progression in naphrobi? This link was established here in a twin and familial cohort study that also included patients with biopsy-proven naphrobi. And what we did at UCSD with Cyril Kossi, that she compared those with advanced fibrosis versus those with early stages of fibrosis, stage 0 to 2, using metagenomic sequencing. And then we also looked at serometabolomics. We identified that 3,4-hydroxyphenylactate, also an aromatic amino acid, was associated with the cause how liver fat might be associated with liver fibrosis. And the shared genetic determination was more than 50%. Therefore, this metabolite is not only linked to differentiation from naphro to normal, but also with those with advanced fibrosis with earlier stages of fibrosis, linking again gut microbial-derived metabolites with disease state cross-sectionally. Here are some preclinical data showing that gut microbiome may promote naphroli by metabolizing fructose. We know that fructose is associated with increased de novo lipogenesis in liver, and this primarily occurs by the classical pathway in the liver, where fructose promotes lipogenesis dependent on ATP citrate lipase here, as shown as a key rate-limiting enzyme for de novo lipogenesis. Now, the novel pathway that was discovered or postulated in this particular paper by Zhao and colleagues, where they noted that dietary fructose is converted by gut microbiome into acetate, as shown here. And then there's a novel pathway where acetate converted into acetyl-CoA, and thereby promoting lipogenesis. And this is dependent on acetyl-CoA enzyme asymptotes. So, now we can see that fructose not only can directly activate lipogenesis in the liver, but it could also be metabolized by gut microbiota, and then in turn produce acetate, and then acetate could be linked with lipogenesis as well. What about gut microbiome in liver disease? There are new data that link that there might be a gut microbiome pattern that may be specific for those patients who have NASH compared to those with NAFL, or those from liver cirrhosis. There is a decrease in the diversity that's seen in patients with advanced NASH. And they also noticed that the type of viruses that are seen in alcoholic liver disease or alcoholic hepatitis are distinct from those with NASH. And some of the details are listed here. Further work needs to be done whether this could be used as a target for therapy in terms of bacteriophages, particularly in those with advanced fibrosis or cirrhosis or decompensated cirrhosis. What about gut microbiome as a signature for detection of advanced fibrosis? This is a question we asked here in this study, where we examined metagenomic signature of gut microbiome in 86 well-characterized patients with biopsy-proven NAFLD at UCSD. These patients were divided into two groups, those with stage 0-2 fibrosis versus those with stage 3-4 fibrosis. What we noticed at the phylum level was that permacutes were higher in fibrosis stage 0-2 and proteobacteria were higher in fibrosis stage 3-4. And this was about three times higher. And then we looked in more details at the species level. And two key species that we identified were Bacteriodes vulgaris and E. coli. And yes, you all might remember when we just about a few slides ago, we talked about E. coli as an important driver of SBP. So this is not just happening in late stages of hepatic decompensation, but this starts happening in terms of changes, subtle changes in gut microbiome and abundance in E. coli even at stage 3-4 fibrosis. These patients did not have any features of decompensation here. Then we were able to develop a metagenomic panel based upon 37 species that could differentiate from earlier stages of fibrosis in patients with NAFLD. And this model had a diagnostic accuracy of about 0.93 in detection of stage 3-4 fibrosis. These data need to be validated in future prospective studies. And this work is currently underway. Now in summarizing gut microbiome alterations that are observed across the spectrum of NAFLD, as you go from NAFLD to NASH to various stages of fibrosis to cirrhosis, we are seeing clear gut microbiome alteration. There is decrease in microbial diversity, higher entry of bacteriaceae and streptococcus, and shift towards more gram-negative microbial environment with increase in proteobacteria and E. coli appearing in stage 3-4 fibrosis. Now how can we link gut microbiome and bacterial metabolites, in this case LPS, with liver fibrosis? LPS can activate TLAG receptor 4, and this leads to release of chemokines and addition molecules that leads to recruitment of Kupfer cells towards the cell itself. BAMBI is a key glycoprotein that's depleted in the setting of NASH and activation of TLR4 by LPS. This leads to activation of TGF-beta, activation of hepatic stellate cell, leading to activation of hepatic stellate cell and fibrogenesis. This pathway is now well known, and several therapeutic strategies are being applied where one could block at the level of BAMBI or block TLR4 to potentially see if we can further improve fibrogenesis as an antifibrotic treatment in patients with NASH and other forms of fibrosis. Here we are looking at, can we identify a first-degree relative who has advanced fibrosis of a patient who has cirrhosis using gut microbiome as a diagnostic test? We have a familial cirrhosis study where we have probands with NAVIRD-D cirrhosis and their first-degree relatives, and then we have several twin as well as family studies where we have twin pair, sub-sib pair, and parent-offspring pair that serve as control in this study. We had 37 first-degree relatives who had advanced fibrosis, and seven of those had advanced fibrosis, and 30 of these 37 did not have advanced fibrosis. Our goal was to develop a model in the probands with NAVIRD-D cirrhosis and validate this in these first-degree relatives. What we found was that the diagnostic accuracy for detection of cirrhosis using just 16S was pretty high. Then we validated that in the first-degree relatives, and we were able to detect with 87% accuracy those who had advanced fibrosis in the first-degree relatives using 27 bacterial features and then adjusting with eight sex and body mass index. We took this work further and developed a metagenomic signature. The idea was to see if we can have a universal signature for cirrhosis. We first applied and validated that in patients in the twin cohort, in the probands with familial cirrhosis cohort, as well as in patients with biopsy-proven cirrhosis within the UCSD cohort. Once we established that, we connected it with the metabolomic and metagenomic basis of this signature, and we were able to identify key nodes within these microbial species that were causing an increase in certain metabolites. Then we were able to take this metagenomic signature in NASH cirrhosis and apply in a cohort, independent cohort in China. That was published by Chin and colleagues in Nature in 2014. Predominantly hepatitis B and alcoholic cirrhosis, we were able to validate the same model that we developed in nephritis cirrhosis in the Chinese cohort, and then also able to validate that in the Italian cohort. We do believe that gut microbiome signature in cirrhosis is so stable that it can detect cirrhosis across etiologies and across geographic areas. What about gut microbiome as a biomarker for treatment response? We know that FGF19, which is a gut hormone that leads to shutting down of bile acid synthesis in the liver, is an important regulator of bile acid metabolism. FGF19, a synthetic FGF19 analog called erdoformin, has been shown to reduce liver fat based upon MRI-PDFF as shown here. What we did was we were able to do 16S using stool samples collected at baseline as well as week 12 in patients treated with treatment with various doses of erdoformin versus placebo. These are the cohorts in the study. We looked at the gut microbiome profile in these patients. There was no difference except for one microbe, which was Vianella. We've previously seen that Vianella is very important in the signature that we looked at in cirrhosis. Here we found that Vianella levels increased by 30-fold. Why is that? We believe that Vianella is a bile acid sensitive microbe. When you shut down bile acid synthesis, this leads to a tremendous increase in Vianella, which was dose-dependent based upon the dose of erdoformin that you used and was also able to predict a treatment response in terms of PDFF reduction. Therefore, you could use Vianella change in stool as a biomarker for PDFF response with erdoformin. This is a new finding and that potentially needs to be translated into other studies using bile acid analogs, as well as FGF19 or other therapies that might be modulating gut microbiome. Key takeaways from this talk on gut microbiome plays an important role in the pathogenesis of hepatic fibrosis, progressional liver disease, and complications of cirrhosis. Gut microbiome-derived metabolites may be involved in development of NASH in progressional liver disease. Gut microbiome-derived signature is a promising biomarker as a non-invasive test for detection of advanced fibrosis or cirrhosis in naphrodite, and gut microbiome may be used as a biomarker of treatment response in NASH-related fibrosis. Thank you, and I look forward to having a panel discussion towards the end. I would like to thank the organizer for this opportunity to present at this meeting, particularly the important aspects as it relates to endocrine factors that can be modifying the management of patients with fatty liver disease. These are my disclosures, largely for drugs that are not still available, but hopefully will for the management of patients with NASH. We're going to cover three aspects, how obesity and diabetes affect naphrodite in a very brief manner. Again, how other endocrine conditions will modify the treatment, and particularly type 2 diabetes. But at the end, I would like to cover three endocrine conditions that are fairly common, thyroid disease, hypogonism, and disorders of growth hormone. So this is a general picture of the role of obesity. As you can observe, you can be born with genes that are not particularly negative in terms of predisposing an individual to obesity, but gaining weight will definitely have a negative impact, not only on the liver, but across the biology of our individuals and lead to metabolic syndrome and cardiovascular disease. The point of this slide is simply for you to observe that type 2 diabetes is clearly a risk factor in which our individuals, no matter how you measure it, are going to be more predisposed to fatty liver disease, and the importance of you as clinicians to keep this in mind as a major risk factor for disease progression and cirrhosis. However, recently we tried to separate obesity by itself from that of diabetes, and you see in blue obesity, and in orange type 2 diabetes. This is the prevalence of steatosis using the fatty liver index. You see that in the overweight range, only 17% of individuals without diabetes have fatty liver disease. As you get into the obesity range, it basically triples. Once your BMI is above 35, three out of four of your patients will have fatty liver disease, and almost 90% once you have higher levels of body mass index. On the other end, having type 2 diabetes makes about half of the individuals with overweight already at risk of NASH having steatosis. Again, this continues to grow, and once you hit a BMI greater than 35, most of your people with diabetes are going to have steatosis, so that's an important thing. Now, the bigger question is how many of them have fibrosis, and again, this is with the FIB4 taking the cutoff of greater than 2.67, about 5% to 10% using this biomarker, but there's a lot of individuals in our clinics in primary care and endocrinology that are in that gray zone. One out of four in our study was found to have potentially some degree of clinically significant fibrosis, so this is an important thing. I know you see the other end, but for us, it's becoming a major, major problem. Now, if we step it up and do elastography, vibration-controlled transient elastography, we can see that roughly, again, 70% overall of individuals who just come for their routine clinics without knowing that they have fatty liver disease have steatosis, and one in five will have fibrosis. Again, liver enzymes are not a very good discriminating factor in our clinics. That's why we are encouraging the use of FIB4, and again, when we broke down the prevalence of individuals that have two to have four, you see that it's affecting roughly one out of six patients, so this is a big number. There are 30 million people with type 2 diabetes in the United States. Again, if you just do the math, that is about four or five million people, so big problem, and big problem because liver enzymes are not a very sensitive way to identify this, so additional screening tests must be done. Finally, again, we can't get too much into this, but the real reason why fatty liver diseases may become even a bigger problem than cirrhosis is because, at least until individuals have F2 or F3 stage fibrosis, I mean, cardiovascular disease remains as a major risk factor for morbidity and mortality, and these are a number of risks amid of all the abnormalities, and I'm sure this group is very, very familiar and mostly related to insulin resistance and lipotoxicity. Now, how does NAFLD really modify diabetes, which is the most common endocrine disorder that we see? Again, as I like to say, type 2 diabetes and NAFLD are like a couple that just get the worst out of each other. Diabetes, from epidemiological studies, seems to promote the progression to cirrhosis and then paracel or carcinoma, and you know it's at least half of the individuals that are in transplant list for liver transplant have diabetes and a significant other is prediabetes, and then when you have fatty liver, we find that they tend to have more atherogenic dyslipidemia, need more medication to get their A1Cs optimized, so it's a lose-lose situation. That is why just about two years ago, the American Diabetes Association began saying, you know, hey, endocrinologists, wake up. It's time that we do something about it, so if at least you have elevated liver enzymes or ketosis, you should begin thinking about NASH and fibrosis, so I think 2022 is going to be a year where this is going to be more aggressively addressed, and hopefully by 2023, we're going to have some hopefully universal screening. Now, why is that? Well, the other thing that we found is that if you don't have diabetes yet, but you have fatty liver disease, your risks increase markedly. This study from South Korea, if you take people without fatty liver disease and you move all the way up to those who had progressive or worse fatty liver disease, you see that as you add severity of fatty liver disease, your risk of diabetes grows, and not surprisingly, treatments that will reverse weight or will reverse insulin resistance prevent the progression to diabetes, but although it's not being looked very carefully if that's because you reverse NAFLD, but just for you to share, those individuals at high risk that are in the box have been studied in many studies now, and you have a two-fold overall greater risk of diabetes if you have NAFLD. One approach that has been tested is lidaglutide at the doses used for the management of obesity. You see large reductions in weight in that upper panel, which are in the range that we can probably improve not only steatosis, but NASH, as you see, greater than 10%, and again, also prevent the progression to diabetes. Studying the semi-glutide probably will lead similar results or with any GLP-1 that reaches those levels of weight loss. Now, interestingly, pioglitazone has also shown to reduce the progression from prediabetes to diabetes, as you can see here, by 70%. There's another study called IRIS that showed the same results, and this is because really what you're doing is making somebody insulin resistant and obese become metabolically healthy, like a lean individual. So this is very important because a lot of your individuals are going to have prediabetes, and I would highly recommend that you do an A1C in all your patients. Dr. Manal Abedmalik is going to be talking later about treatment, so we are going to leave her, expand on the role of pioglitazone. I've been involved in some of these studies. This is the longest study, three years of follow-up with quite positive effects on steatopatitis and some modest effects on fibrosis. But again, we think that this is important that primary care doctors and practicing gastroenterologists, hepatologists know it, because the choice now is to eventually, if you identify somebody with diabetes, particularly, and you have to choose between using metformin or a sulfonylurea that don't have an effect on NASH. I mean, you may choose an agent like a GLP-1 receptor agonist, particularly the data is positive with semaglutide, or pioglitazone, which costs less than $5 a month to achieve a dual goal of improving glycemic control, but also reversing NASH and hopefully the progression of fibrosis. Last but not least, there's a clinical care pathway just came out in September, done with the initiative of AGA and Dr. Canwell, who invited me as an endocrinologist, and Dr. Sherbrooke as a primary care doctor, and a number of very prestigious colleagues who came up with a clinical care pathway. This is largely to help us, the non-hepatologists, to manage these patients. Again, dividing them into low risk, intermediate risk, or indeterminate risk, and high risk, and trying to emphasize the need for lifestyle changes. Again, cardiovascular disease risk reduction. Do not stop the statins. These patients have high cardiovascular risk and they're fairly safe. More aggressive lifestyle changes, and again, awaiting FDA approved pharmacotherapy, but again, particularly in those with diabetes, even perhaps with pre-diabetes, as I mentioned, pioglitazone, very inexpensive drug, GLP-1s, expensive drugs, but they both have shown pretty convincingly to help our patients with reversal of statin hepatitis, resolution of NASH, and in some instances, the reduction of the progression to cirrhosis. With this, the take-home message is that weight gain will lead to insulin resistance, type 2 diabetes, and cardiometabolic risk. This is reversed by any ways that you promote weight loss, or can be reversed even with mild weight gain with P-bargamma agonist, as shown with pioglitazone, and in the native study with Laniferinor, that although you may gain 2% or 3% body weight, you achieve the same metabolic and cardiovascular targets as with weight loss. With that, remember that your approaches today from a metabolic and even NASH resolution perspective, either you reduce the adipose tissue mass or you reverse the adipose tissue dysfunction with a drug as pioglitazone, and hopefully in the future, compounds like Laniferinor. You reduce the mass or you change its biology, and I think that's a pretty practical approach for you to remember moving forward. Okay, so the final 10 minutes, I want to discuss three common conditions you're going to find in your clinic, and I want to tell you what the contribution is overall to NASH. There are a number of endocrine disorders. We're not going to make you an endocrinologist, although I have become a gastroendocrinologist, I have to say. But again, hypothyroidism, very exciting field. It has very involved almost with any pathway. Thyroid-responsive elements are broadly distributed in the body and particularly in the liver, altering a number of pathways. Again, there are deodorization and other pathways that are turned on and off depending on your metabolic status, whether you're fasting or not. And again, very important to know that in primary hypothyroidism, there's abnormalities metabolically and particularly in the liver that are very, very important. Now, when you look at the studies, the studies have a more mixed, this is a nice review in thyroid by Dr. Mantovani. Again, except these two studies, others have been a little bit less convincing. So we looked into this recently and we measured liver fat by magnetic resonance spectroscopy. As you become more hypothyroid, this is the TSH that's missing. You're suggesting that you're developing hypothyroidism and that promotes steatosis. So hypothyroidism appears to promote steatosis. However, in these individuals with biopsies, we did not see that being more or subclinically hypothyroid or frankly hypothyroid had anything to do with other features of NASH, particularly, definitely not with fibrosis. So I think the relationship is modest and weakens considerably when you include the usual suspects, such as obesity, diabetes, or insulin resistance. That doesn't mean that a thyroid responsive element like smeritum or others that are in development would not be beneficial. It just says that it is a complex biology and at least at the clinical level, if your patient is hypothyroid, this would probably not be a major risk factor or would not explain the severity of their liver disease. Let's shift to sex hormones. Again, without trying to make you become an endocrinologist, I wouldn't dare to do that. Estrogens and androgens have huge effects on glucose and lipid metabolism at different levels. Again, we looked at this very carefully in individuals that were biopsied, I think about 200 patients in this study. Again, when you normalize for not divided between non-obese and obese, you see that testosterone levels drop with obesity and they're very low eventually in individuals with NASH that are very insulin resistant. And when we looked specifically at histology, we did not see that, although there was a trend for steatosis with hypogonism, we did not see that this had a major impact on other features of liver histology. So overall, when you factor in, again, the big suspects, insulin resistance, obesity, or diabetes, the impact of hypogonism in males weakens. And the same, we did not study, and this is with, we've looked at the same for post-menopausal women. And again, steatosis trends to go upwards, but it's not a major factor for other features. And as you suspect, the total testosterone level has an inverse relationship with obesity, some degree with liver fat, but also with insulin resistance. So I would have to conclude that it is a modest factor overruled by obesity and insulin resistance. Last, what about growth hormone? Again, many, many papers, very confusing literature. Yes, there's a good rationale. Many mechanisms can be involved in growth hormone metabolism regarding fatty liver disease. This is a nice summary, fairly recent. Again, some effect favoring, I've seen higher, I've seen abnormal IGF-1 levels in individuals with fatty liver disease. We also looked at this carefully. Again, all of these relationships weaken when you adjust for obesity, diabetes, and insulin resistance. But we looked at it more carefully. We looked at FGF-21. FGF-21 is the mediator. Growth hormone is secreted and activates in the liver, the production of fibroblast growth factor 21. And we do see that there's like an FGF-21 resistance in NASH. Levels are higher. Particularly, we saw this at the level of skeletal muscle. I spared you more details of this study, but I did want to share with you the histology. Again, not a big effect in terms on steatosis, but a major effect on necrosis and ballooning and on inflammation. We see paradoxically higher FGF-21 levels in these individuals. And when you look at fibrosis, same trend. Clearly, there's a biology very poorly understood in humans in which there's not a major effect of growth hormone deficiency by itself, but some sort of FGF-21 resistance with higher levels. And that, again, appears to justify the intensive research that is being done in the field of FGF-21 analogs. There are many compounds. Last time I counted, there were like seven in different stages of development and with different effects depending on their target. So to finish, let's just wrap up all this information given to you in 20 minutes to say that obesity and type 2 diabetes are the two major endocrine drivers of fatty liver disease and progression. That's nothing new, but what is new is that you should be always doing hemoglobin A1C in your patients. Very, very important. NAFLD promotes the development of type 2 diabetes. So your approaches in a simplistic way where either you change the mass of adiposity, you promote a reduction with weight loss, or if you can defeat obesity, you can change its biology, making your individuals have the adipose tissue of a lean person. And finally, other endocrine conditions are important. So if you have somebody with fatigue, you can ask them to have an endocrine evaluation, but the role on NASH is rather limited. So with that, thank you for your time and I'm excited to eventually address any questions or you can email me and I'll be happy to help you. Have a great day. Well, it's a pleasure to be here speaking with you all. I'll be talking about cardiovascular disease and non-alcoholic fatty liver disease. So as you all know, people with NAFLD have a decreased 10-year survival compared to the general population, and CVD or cardiovascular disease-related mortality accounts for up to 25% in those with non-alcoholic fatty liver disease. The question remains though, is NAFLD an independent risk factor for cardiovascular disease or is it the concurrent metabolic diseases that we see related to NAFLD, such as diabetes, dyslipidemia, and hypertension lead to the increase in cardiovascular disease that we see? Well, one study did recently re-evaluate this. This is a study by Targer's group that did a meta-analysis on 36 longitudinal studies looking at nearly 6 million individuals with nearly 100,000 fatal and non-fatal cardiovascular disease events over a median follow-up of 6.5 years. They found that NAFLD was independently associated with increased fatal and non-fatal cardiovascular disease events with a pooled random effects hazard ratio of 1.4, and that this risk of fatal and non-fatal cardiovascular disease events increased as fibrosis stage also increased. However, we do know that some of the studies that were included in this meta-analysis and studies that have been done on cardiovascular disease and NAFLD have understandable limitations. Some lack assessments of baseline cardiovascular disease and coronary heart disease. Some lack complete adjudication for cardiovascular disease risk factors because they are not all conducted prospectively. Most lack adjudication by blinded cardiologists to cardiovascular disease outcomes. What our group tried to do to get at this answer of whether NAFLD is an independent risk factor for cardiovascular disease is look in something called the PROMIS trial, which I'll tell you more about. The objective of this was to compare rates of incident rather than prevalent, but incident major adverse cardiovascular disease events or MACE by steatosis status by controlling for cardiovascular disease risk factors, baseline atherosclerotic disease burden, and using adjudicated CBD outcomes. MACE, as it's often defined, is death, myocardial infarction, or unstable angina requiring hospitalization. The study trial was a nested cohort study from the PROMIS trial. The PROMIS trial was a study of over 10,000 individuals with stable chest pain who were randomized either to undergo functional testing like stress testing or coronary artery CTA. Of those, 4,966 had a CTA and 3,756 also as part of their CTA had the liver and the spleen captured and were able to be assessed for steatosis, which was found in what we would expect about 25.5% compared to those with a normal liver and 74.5%. We found that steatosis at baseline was associated with a higher prevalence, not unsurprisingly, of obesity, the metabolic syndrome, type 2 diabetes, and a higher cardiovascular disease risk burden that's assessed by the atherosclerotic cardiovascular disease or ASCBD risk score. These patients had a higher burden of cardiovascular disease risk factors than those without steatosis. This is a score I'll come back to. It's important to know it's a fairly comprehensive score that can be calculated online. It's a composite score of age, gender, race, total cholesterol, the presence of hypertension, systolic blood pressure, tobacco use, and type 2 diabetes. At baseline, differences in atherosclerotic burden by steatosis were found in the atherosclerotic burden by steatosis were small. Despite these differences in risk factors, the baseline differences were small. There were no differences in the prevalence of high-risk plaques, calcified or non-calcified plaques. Although steatosis was associated with a higher mean Lehman score, which is a total score of obstructive and non-obstructive burden in the cardiovascular, in the coronary arteries, and Agassin score or coronary artery calcium score. Interestingly, when we turn to incident disease, we had a median follow-up of 25.5 months and the overall rate of MACE, major adverse cardiovascular events, was 3.1 or 115. Baseline steatosis was associated with significantly higher rates of MACE at 4.4% versus 2.6 with a hazard ratio of 1.72 and a p-value of 0.007, as you can see here on the graph to your right. This was significant after adjusting for all relevant covariates that were prospectively collected and after adjustment for all CT measurements of baseline plaque and baseline stenosis. It's not the fact that patients with steatosis have more baseline atherosclerotic disease that led to higher MACE, but in fact, even after adjusting for those, those with steatosis had higher incident rates of MACE. Further, we can see here that as you add steatosis to different degrees of non-obstructive and obstructive CAD, so too does the hazard ratio for MACE increase. In the green lines, we'll see patients with no CAD at baseline. When you add hepatic steatosis to that, the hazard ratio increases. In the yellow lines, those with non-obstructive CAD in the dotted yellow line have a lower rate of MACE than those who have steatosis plus non-obstructive CAD. In the red lines, you can see that those in the red straight line who have significant steatosis and significant stenosis also have significantly higher rates of MACE than those without hepatic steatosis. What we found here was that baseline hepatic steatosis was associated with a 70 percent increased risk of major adverse cardiovascular disease events. This increased risk was independent of traditional risk factors and independent of the present of extent of baseline CAD, including obstructive CAD and measures of plaque burden. However, we wanted to look further into this and look to see if there are traditional cardiovascular disease risk factors in NAFLD and are there unique risk factors for patients with radiographic NAFLD or biopsy-proven NAFLD. Again, we turned to the PROMIS study and the nested cohort study, but we limited to those in the green triangle, those just with steatosis, and again, looked at MACE to try to understand why do some people with steatosis develop cardiovascular disease versus some who don't and what are the risk factors that might explain this. We found that not surprisingly, risk factors for significant CAD at baseline, so this is prevalent CAD, included age and male gender, as well as diabetes, hyperlipidemia, and smoking, something that I think we pay not enough attention to in our liver clinics. We also found that the ASCVD risk score, the score that I mentioned before, did predict the presence of prevalent cardiovascular disease with the risk going up 5% by every one-point increase in ASCVD score, suggesting that the ASCVD risk score is a reasonable way for us to risk stratify our patients for prevalent CAD. We also identified other known risk factors, including LDL, total risk of cardiovascular total cholesterol, triglyceride, and ApoB levels. NT, ProB, and P, and high-sensitivity troponin were also predictors of prevalent CAD. What about risk factors for incident disease or incident major adverse cardiovascular disease events? Well, interestingly, because of the high prevalence, we believe, of most of these conditions in our patients with steatosis, we didn't see that things like hypertension and diabetes were independent risk factors. What we saw were the strongest risk factors were, again, the ASCVD score with a three-percentage increase in MACE for every one-point increase in ASCVD risk score, and that sedentary lifestyle was a significant risk factor for major adverse cardiovascular disease events in NAFLD with a hazard ratio of 2.57. CAD severity and burden in NAFLD was also associated with traditional risk factors, including modifiable risk factors, NT, ProB, and P, and high-sensitivity troponin, which is something that's being seen in the non-NAFLD population. But most importantly, what we found is that the ASCVD risk score can be used to predict MACE in our patients with NAFLD, and I'll show you how the ASCVD risk score can be used to determine whether a patient would benefit from statin therapy and what type of therapy, and that sedentary lifestyle is really an important and modifiable risk factor in our patients with NAFLD for major adverse cardiovascular disease events. However, we're still limited in these studies by a lack of histology, and we can't determine what impacts the components of NAFLD and NASH, as well as fibrosis, have on CVD risk. We looked at patients with biopsy-proven NAFLD and long-term follow-up to identify predictors of incident CVD. We looked at 285 adults with biopsy-proven NAFLD without CVD from the MGH NAFLD cohort, and our primary outcome was incident cardiovascular disease, defined as a new diagnosis of coronary artery disease, heart failure, peripheral vascular disease, CVA, cerebrovascular accident, or TIA, or MACE. Patients were followed until their first incident CVD event, liver transplant, death, or the end of follow-up, and were followed for a mean of 5.2 years. Incident cardiovascular disease events occurred in 26 or 9.1% of the cohort over this time. We found that similarly to those in the PROMIS study, smoking was a very important predictor of incident cardiovascular disease, as was low albumin, suggesting perhaps that advanced liver disease may be associated with cardiovascular disease. Advanced fibrosis, so fibrosis stage 3 or 4 out of 4, was one of the highest predictors for incident cardiovascular disease with a hazard ratio of 2.86, and these were after adjustment for relevant covariance. Ceatosis, hepatocyte ballooning, lobular inflammation, or the presence of NASH were not independent predictors of incident CVD. On histology, only advanced fibrosis was a predictor. You can see here that same information presented graphically, that advanced fibrosis in the blue had a significantly higher rate of incident CAD. Interestingly, if you look on the left, when we substituted advanced fibrosis histologically for the NAFLD fibrosis score in our models, a NAFLD fibrosis score greater than 0.676 was also a predictor of incident cardiovascular disease, suggesting that even looking at our patients noninvasively, we would be able to re-stratify them for cardiovascular disease. In adults with advanced fibrosis at baseline, they had significantly higher rates of incident cardiovascular disease compared to those with fibrosis stage 0-2, and among those with biopsy proven NASH, smoking and advanced fibrosis were the strongest predictors of incident CVD. What do we do about this? Well, in adults with NAFLD, we recently published guidelines or comments to help provide guidance on the management of cardiovascular disease in those with NAFLD in diabetes and metabolism in 2021. Of course, the foundation of all treatment of NAFLD with or without cardiovascular disease is lifestyle modification with physical activity 150-200 minutes per week following a Mediterranean diet and limiting saturated fats, red and processed meats, refined carbohydrates, and sugar-sweetened beverages. For patients with dyslipidemia, and we'll talk about this a little bit more, they really should be considered for statin treatment and perhaps preferentially lipophilic statins which may have an additional benefit in HCC risk. Hypertension starts out if the hypertension is in stage 1 with non-pharmacologic treatments and lifestyle intervention, but if unsuccessful or if it's higher, then we recommend ACE inhibitors or angiotensin receptor blockers as they may have a possible antifibrotic effect that is not seen with other antihypertensive medications. Treating patients and advising tobacco use or referring them for interventional or pharmacotherapy is key. Treating type 2 diabetes and perhaps using SGLT2s or GLP1s that also have Nafl benefit should be considered. For those with obesity, the goal is a total of 10 percent total body weight loss, a hypercaloric diet of about 500-1000 calories less than baseline and considering pharmacologic therapy or bariatric surgery. In our take-home points, hepatic steatosis is associated with a 70 percent increased risk of MACE independent of traditional CBD risk factors. Among those with Nafl, baseline CAD and severity and burden in Nafl, it's associated with traditional risk factors as well as ASCVD score and sedentary lifestyle predict MACE in Nafl, and advanced fibrosis is associated with incident CBD. It's important to remember in our patients with Nafl, that those with advanced fibrosis have higher risk of incident CBD, so we need to be paying even closer attention to their cardiovascular risk factors and sedentary lifestyle, so encouraging aerobic exercise or resistance training is really key to help decreasing the risk of MACE in these patients. We also have shown that the ASCVD risk calculator is beneficial and can predict outcomes in incident MACE in our patients with Nafl, and this is a free online calculator that allows you to enter in demographic and lab data, as you can see to the left, of patients, and it will give you the 10-year and lifetime risk of ASCVD. Then it will also give you recommendations, as you can see on the left, recommendations about statin initiation and the intensity of the statin that should be used. Finally, newer guidelines are available from hypertension, the journal Hypertension, about management of established diagnosis of hypertension and lifestyle advice is key for patients with grade 1 hypertension, so blood pressure 140-159 over 90-99. They first are considered to undergo immediate treatment if they have high-risk features or if they have failed lifestyle intervention in the past. For those, if so, then they go on to drug treatment. If not, they can be considered for lifestyle intervention and for patients with higher blood pressure, they are recommended for immediate start with blood pressure medications, and as we recommend, ACER, ARVs would be our preference for starting in these patients. We do recommend working closely with the cardiologist to help manage both the dyslipidemia and the hypertension. Treatment calendar calculators, including the ASCVD calculator, can determine when and how much lipid-lowering medication to start. Hypertension guidelines provide information on when to start pharmacotherapy for hypertension and for all of our patients, but especially those with NAFLD and advanced fibrosis, lifestyle intervention with weight loss, smoking cessation, and Mediterranean diet, and an emphasis on physical activity to overcome that sedentary lifestyle should be recommended to all. With that, I'm happy to take any questions. Thank you so much for your time. Good afternoon, everyone. It's a pleasure for me to talk to you. The title of the talk I'm going to give is hepatocellular cancer in non-sorotic NASH, a mountain or a molehill. My name is Helen Reeves. I'm the professor of liver cancer in Newcastle University in the UK. I did a PhD a number of years ago in liver fibrosis looking at cell signaling pathways, and then subsequently had an AASLD fellowship, and I worked with Scott Friedman at Mount Sinai. This is the first time I've talked at the postgraduate course for the AASLD, and it's a real pleasure, so thank you for the invitation. This is my disclosure slide. Today we're talking about the cancer risk in liver disease. What we can see here is the normal histological pattern of a person's liver without injury. Here's a portal tract where everything comes in via the hepatic artery or the portal vein. You can imagine all that blood going through these sinusoids between the sheets of liver cells to the central vein, where it will leave the liver. Now, if we have any kind of liver injury, there will be an inflammatory response where you can see immune cells infiltrating the liver, perhaps expansion of these portal tracts, fibrosis around the portal tracts, and fibrosis around the central veins. When we talk about bridging fibrosis, it's where we've got that link between the two or between adjacent portal tracts. We end up with this appearance here, where a person has cirrhosis with nodules of hepatocytes that are regenerating and the injury continues. It's simpler to understand how you might have injury happening here, where there's a mutation in a gene that's perpetuated in the presence of continued proliferation as a regenerative response. If you have a patient that's got fatty liver disease underlying that process, why wouldn't they also have the same risk? But what we're talking about is patients like this, where here you have somebody with inflammation, some ballooning degeneration, in fact, but without that nodular regeneration. Here, even a patient with simple steatosis, without that level of injury and inflammation, and these patients may also develop cancers. But what truly is their risk? We're going to look at some of the epidemiological evidence. We're going to also look at this population and consider whether or not we should be screening them or performing surveillance for liver cancer. If we're going to do that, what tool are we going to use? If we think that's hard, is there a way that we could identify patients who are at greater risk of developing hepatocellular cancer? We'll also just touch on what are actually the current recommendations. What is the evidence? Now, until recently, we haven't had that much evidence. It's been case reports, probably with bias. In 2014, we had this overview of what was available in the literature published. The suggestion here was that patients with cirrhosis over a 20-year period had got a risk of between four and 27 percent of developing a cancer. For patients with NASH, that was smaller, anything depending on the study between naught and 2.8 percent, and those with simple steatosis, lower still at naught to 0.5 percent. It was looking like it was low, but it was recognized that more evidence was needed. Now, we do have quite big epidemiological studies. I'm not going into the details of this, but here's one from Gastroenterology in 2018, reporting that it is 10.6 per 1,000 patient years as a risk for a patient with naffle cirrhosis. If you take away the cirrhosis, that is much lower at 0.08 per 1,000 years. In the same year published in the Journal of Hepatology, we had this very large collaborative paper looking at modeling of naffle disease burden across the continents, and predicting what the likelihood of problems are going to be for the next 15 years. In that paper, they looked at the disease stage annual transition rates depending on what was the level of injury in the patient with a biopsy at that time. This was looking at progression to cirrhosis and also progression to cancer. Here's the person that has cirrhosis. What is their annual rate of progression to HCC? Perhaps not quite as high as we would expect, but 0.34 percent. What was clear was that the less fibrosis you had, the lower your risk. Down here, we're going to have patients with NASH who maybe have a little bit of fibrosis, and their rate is down here at 0.011 percent or 0.004. That means that what you're looking at for an individual person is a very small risk. You would think that that isn't a mountain, it's a molehill. But before we can really address the problem, we have to ask how big is the at-risk population before we can decide if it's an economic burden for the people that look after patients with liver cancer. How common is NASH? There's a lot being talked about at this postgraduate course on this, but this is a global chart that we're familiar with now showing the prevalence of obesity. The darker color blue in your nation, the higher your level of obesity. We know from the WHO that worldwide obesity has trebled since 1975. There, the implication is that we have nearly two billion adults that are overweight with 13 percent of them being obese. We're obviously talking about a very large at-risk population, and from quite a lot of epidemiological studies that have been performed over the last decade or so, we suspect that the prevalence of NAFLD is around 25 percent. Is it a mountain or a molehill? Globally, we think there are 650 million obese adults. If we think 25 percent have got NAFLD, that's quite a lot of patients there and their risks of cancer. Even if we take those very low levels of risk, that would mean that we're seeing quite large numbers of patients globally that will be presenting possibly with a liver cancer. If we just look at the USA with a population of 328 million, and we look at this prevalence of NAFLD at 10-20 percent, 2-5 percent with NASH, relatively conservative estimates. But we're perhaps looking at between 72,000 and 180,000 new cases per year with non-sorotic NASH HCC. Although individual risk is small, this is actually a mountain in terms of the impact that it can have on our society. In terms of morbidity, economic burden, and mortality. What can we actually do about this? Can we detect cancers by surveillance? If we think the risk is that small and the population at risk is so high, that might be deemed as similar to finding a needle in a haystack. We need to know the tools. This is my husband and my dog. If we were waiting for them to find a needle in one of those haystacks, we'd be waiting a very long time. What do we know about surveillance and the tools that we have for detecting liver cancer? First of all, what is surveillance? We know this is the periodic application of a diagnostic test to individuals at risk with the aim of reducing mortality. To reduce mortality, we have to detect the cancer at an early stage, where cost-effectiveness is going to depend on how common the disease is in the target population. Modeling suggests that it needs to be there in about one and a half percent of patients. We need to have a good test that's not too expensive, and it needs to be an acceptable test for the target population. If you find the thing you're looking for, you have to have a treatment that's going to make a difference that your patient is fit enough for. There's been a lot of debate over the years about whether or not we should be performing surveillance in patients with cirrhosis. It's not ethical to do a randomized controlled trial where you know there is a level of risk, however big or small it is, for getting liver cancer. We look at cohort studies rather than randomized controlled trials. But I think it is widely accepted now that this is a worthwhile thing to do. Here's quite a recent meta-analysis showing that for the tools for a cirrhotic patient, the sensitivity of ultrasound alone is in the region of 45 percent. The sensitivity of ultrasound and alpha-beta protein combined is 63 percent. If you're going to do this, probably you should be using the combination of those tools in patients with cirrhosis. But what about those without cirrhosis? We know that the incidence of liver cancers is not as high as one and a half percent per year. But surveillance in a group of patients who don't have cirrhosis could possibly be acceptable with a lower risk, because even if somebody had a large tumor, a more advanced tumor, they might still be amenable to curative treatment by resection or tolerate palliative therapies better. Modeling would suggest you might have a lower incidence and still have cost-effectiveness. But the problem is, as we've seen from the epidemiological data, the risk is nowhere near that high. So we're not going to be approaching cost-effectiveness with the tools that we have. So can we stratify risk? Can we identify those patients that have got a higher risk to focus our efforts on them? So what do we know about risk elevators and possibly the mechanisms, the biological mechanisms that would guide us to the tools of the future? This graph is looking at patients presenting in my center in the north of England. And this is the numbers of patients referred to our multidisciplinary team up until 2010. This has actually doubled now in 2020. But what I'm pointing out is this orange bar. So this is the patients with NAFLD-HCC, not actually recognized in the year 2000, but by 2010, actually the commonest cause of liver cancer on our unit. We know that those patients are older than the patients with viral hepatitis or alcohol-related cancers, but we're able to actually characterize those that are developing liver cancers without cirrhosis a little bit more. What do we know about them? They make up 20 to 30% of the NAFLD-HCC patients. They're predominantly male. Their median age is older still. So there's no point doing any surveillance on 60-year-olds without any fibrosis. You're never gonna find a cancer. We also know that two-thirds of them have type 2 diabetes. So if we're thinking about risk stratifies, it's probably age, gender, and the presence of type 2 diabetes, looking at that. What else? What about the genetic risk? So there's been a lot of talk over the years about these polymorphisms, single nucleotide polymorphisms, in PNPLA3 and TM6SF2. And these are genes that contribute to the regulation of fat processing and storage within the liver. And simply put, the more fat you have, the more injury you have, the more of a healing response with fibrosis, cirrhosis, and regeneration you're gonna get, and then maybe you'll get more HCC. But is there actually a risk attributed to those genes in patients without cirrhosis? And the reality is that it's probably a very, very, very small one. The major risk, cancer risk, attributed to them, I think, is as a result of the cirrhosis. But we're starting to appreciate the importance of the immune environment. And so these patients, regardless of whether or not they've got cirrhosis, have a modified immune environment where, for some reason, they're more likely, some of them, to develop cancers. This is a paper my group has recently published looking at the role of a SNP in PDCD1. And this is the gene that encodes the T-cell receptor PD1, the checkpoint receptor. And so we're developing this idea that in cirrhosis, there are these genes that promote the fatty change and progression to cirrhosis that are important. And probably in the patients without significant fibrosis, maybe it's polymorphisms in the genes regulating the immune environment that might be more important. What about the biology? There's been a number of animal models published over the years. This is one recently published by my group where we've developed a model in which mice develop fatty livers with NASH and cancers, but in the absence of cirrhosis and in the absence of giving them an additional mutagen. So we think that this is reflecting the human condition a little bit more accurately than some of the mouse models. And some of these findings are not unexpected. They're male mice, the fatter ones have bigger livers and they get more tumors. And elevated blood glucose was also associated with more tumors. And this was associated with a larger tumor size. But what else could we find out from this mouse model and here, because it was a mouse model that was more similar in terms of the natural background to humans, we also had the ancillary features of NASH. So we're looking at things like not just ballooning degeneration and lobular inflammation, malarie-dength bodies, pigmented cut for cells, but it was these guys here that were associated with the development of cancers, the lipogranulomas. So this is macrophages moving into free fat from ruptured hepatocytes, creating this altered immune response, most likely upstream of manipulating T cells that were driving actually a proliferative environment. And this was occurring in the absence, really, or with minimal ballooning or DNA damage and fibrosis. Something else was driving that proliferative response associated with the development of cancer or the risk of cancer. And we phenotyped these macrophages a little bit more and showed that they were acquiring a CD44 expressing phenotype. So we're developing an idea here where you have a proliferative response, even if you don't have cirrhosis, this is an overweight people, often men, where there's an elevated fasting blood glucose, more fat in their liver, and it's this immune response to fat, perhaps governed by lipogranulomas, which change in character, which is upstream of a T cell environment, which can be immunosuppressive where perhaps genetic variability plays a role. What about the guidelines that are telling us what to do at the moment? The ASLD guidelines advising us that all patients with liver cirrhosis, except those that are child's puberty C, should have surveillance, see if they're on a transplant waiting list, and the recommendation is an ultrasound and AFP. The easel guidelines are similar, but there is a stratification score for patients with viral hepatitis B, based on platelets in keeping with portal hypertension, age, gender, and your hepatitis B DNA level. The easel guidelines also suggest you to consider surveillance in all non-cirrhotic patients who have F3 fibrosis, falling short of cirrhosis. That's a lot of patients with NAFLD, and because we probably can't identify those patients particularly well, I don't know that this has had that much of an impact as yet. The only other feature about the easel guidelines to mention is that the recommendation is for ultrasound presently, rather than a combination with alpha-beta protein. So might there be a stratification tool that we could adopt for patients with NASH? That's a little bit better. We don't know about platelets. We haven't looked. Age probably is a factor. Gender probably is a factor. Other factors for these patients might be family history. Maybe we'll be developing polygenic risk scores. Co-factors would include type 2 diabetes, perhaps moderate alcohol smoking, drugs that are protective, metformin, insulin that might promote risk. So there's a lot that we still need to do here to work out what is the best stratification regime. And other biopsy information that you might look for in your patient if they've had a biopsy might be iron globules in keeping with alpha-1 antitrypsin deficiency. I don't know that we're going to be looking for CD44 positive macrophages, but the CHI-67 score as a proliferative index might be something that could be applied because it is used in histopathology scores for other diseases like cancers. Different tests, the serum test, many of us know about the GALAD test, which is a stratification which uses a number of different features, gender, age, alpha-beta protein, and des-gamma-carboxyprothrombin. This also includes this modified protein, AFPL3, which is more cancer specific. And so we may see this coming along as a surveillance tool in the not too distant future. So this is my takeaway message. Looking at the green box first, we know that NASH HCC, the individual risk is very, very small for a patient in front of you. So even though it's small, this is a mountain in terms of the economic burden and mortality. The current surveillance tools that we have are not cost effective in this population. If we're going to actually perform surveillance, this might be in those that have got more advanced fibrosis. And if we're going to do this, if we come over here, if we were gonna consider surveillance six monthly, it might be if this was a fit person for a start who had F3 fibrosis, you would be more likely to do it if they were male, if they were older, if they have type two diabetes, and perhaps if they had a biopsy, which showed some features that we know elevates risk. And on the evidence that we have available, we would probably use ultrasound, which is possibly better in non-sclerotic patients because it's not so difficult, easier to see a nodule, but in combination with alpha beta protein would be my recommendation. And so that just leads me to thank Newcastle University, my employer, Newcastle-Upon-Tyne Hospitals NHS Foundation Trust, which is where I look after all of my patients and the funders for the research program in Newcastle Cancer Research UK. So thank you very much, everybody, for listening. Hello, I'd like to thank the program organizers, Dr. DeLev and Dr. Ranello for the opportunity to speak today. I'll be talking about the management of non-alcoholic fatty liver disease, what we can achieve with available therapies. My name is Manal Abdul-Malik. I'm professor of medicine and director of the NAFLD Clinical Research Program at Duke University. Here are my disclosures. So in the next 20 minutes, I'd like to review those pharmacotherapies which have potential to decrease liver-related outcomes, discuss the known histologic benefit of repurposed medications and NASH, and to review the effects of bariatric endoscopy and surgery on NASH. We've all come to recognize that non-alcoholic fatty liver disease and steatohepatitis is a dynamic process that can progress and regress, and that the patients of most concern and risk are those patients with steatohepatitis and hepatic fibrosis. Lifestyle modification is the backbone of therapy for all patients. And in the absence of an FDA-approved drug, targeted pharmacotherapy for FDA-approved indications targeting comorbidities of NAFLD and NASH is all we have in our armamentarium. And when medical treatment is unsuccessful, there is consideration of bariatric endoscopy, surgery, or even referral for clinical trials. Pharmacologic therapy is discussed in the ASLD guidelines, the ESL guidelines, and the Asian Pacific guidelines, with the ASLD taking the vantage points that biopsy-proven NASH and those with fibrosis have an indication for pharmacologic treatments. The ESL guidelines also states that pharmacotherapy is indicated for those patients with NASH, particularly with those that have significant fibrosis, but makes the stand that those patients with less severe disease who are at high risk for disease progression, such as those patients with diabetes, who are enriched with metabolic syndrome or have high necroinflammatory grades could also be candidates for pharmacotherapy. The Asian Pacific guidelines states that patients without steatohepatitis or fibrosis should receive no pharmacotherapy and should be guided on lifestyle interventions and healthy diet. Now, these guidelines underscore the role of weight loss, whether that be lifestyle modification, including dietary change or exercise, but that weight loss is the backbone of treatment for NAFLD and NASH. With diet, there is limited data on macronutrients, but that energy deficit of 500 to 1,000 calories can induce potentially a weight loss goal of 500 to 1,000 grams per week. We exclude those dietary components which have been associated with NASH, such as fructose or processed foods that are high in hydrogenated fats and encourage a Mediterranean-style diet. A weight loss of seven to 10% is the target for lifestyle interventions to improve not only NASH, but fibrosis. There is little data on the effect of exercise on liver histology, but all guidelines recommend a tailored approach to a patient's preference. And bariatric surgery has been demonstrated to reduce liver fat, improve histologic features of NASH and fibrosis, but all the guidelines recommended an individualized decision, particularly in those patients who have advanced hepatic fibrosis or cirrhosis. So the treatment of obesity is the foundation for the treatment of NAFLD and NASH. A 5% reduction in BMI translates into a 25% relative reduction in liver fat by MRI-PDFF. In a systematic review of 23 trials, weight reduction of four to 10% consistently improves liver fat content and liver aminotransferases. And in a recent meta-analysis of 26 studies, as you can see, consistently weight loss improves liver aminotransferases and weight loss consistently improves liver fat content. In a paired liver biopsy study of 261 patients with biopsy-proven NASH who underwent a 52-week lifestyle modification, increasing thresholds of weight loss was associated with increased improvement not only of liver fat, but at thresholds greater than 7% resolution of steatohepatitis and at thresholds beyond 10%, even an improvement in liver fibrosis in 45% of patients. The unfortunate reality is only a minority of our patients can achieve these thresholds of weight loss of 10% or higher, and even a smaller proportion can sustain such thresholds beyond a year. Exercise as a treatment of NASH has been shown not only to improve liver fat, but ALT. And in a meta-analysis of 28 randomized trials of exercise-based interventions in patients with NAFLD and underlying metabolic disorders, exercise reduced intra-hepatic lipid content and also reduced serum ALT. And so the majority of studies suggested a benefit to exercise in the treatment of NAFLD. And I'd love to point out that there is data that states that exercise can retard hepatocarcinogenesis in this beautiful animal study of the FOSFOS mouse, which is an insulin-resistant diabetic mouse that when fed a high-fat diet develops steatohepatitis antipatic fibrosis, that when these mice were introduced to our carcinogen and kept sedentary, that the FOSFOS mouse developed hepatocellular carcinoma. However, if they were subjected to routine physical activity and exercise, not only did the tumor burden of liver cancer decrease, but there were fewer foci in the liver. And this was associated with a reduction in weight and an improvement in insulin resistance. And this is translated into epidemiologic studies that suggest that vigorous physical activity more than five days a week decreases the relative risk of liver cancer by 44%. So it's imperative that our patients exercise. But when combined with data suggests that exercise improves portal pressure, it becomes even more notable that we have to integrate this into our clinical practice. In a study of 50 patients with compensated cirrhosis, 92% who were child's A, 72% of whom had clinically significant portal hypertension of greater than 10 millimeters of mercury, these patients were overweight and obese. Intensive lifestyle intervention was performed. And in fact, despite the presence of cirrhosis, 52% of patients were able to lose at least 5% or more of their body weight. And 42% had a reduction of at least 10% in hepatic main pressure gradient. And this reduction correlated with the category change in body weight. So the higher the percent of weight loss, the higher the probability of decreasing portal pressure. Now, currently we don't have available FDA approved therapies for NASH. However, there are therapies that can potentially be utilized off-label that may have an improvement in NASH. Going back to the guidelines, I'd like to point out that metformin, while it has been studied in several small studies, is not recommended as a treatment for NASH. Vitamin E is recommended in the ASLD guidelines in non-diabetic patients with biopsy proven NASH. And the ESL guidelines also recommends vitamin E at a dose of 800 international units per day. The Asian Pacific suggests that there's insufficient evidence as of yet. Pyoglitazone is recommended in all the guidances with the recommendation to be in patients with or without diabetes. But the ESL guidelines and the Asian Pacific guidelines restricting consideration of pyoglitazone for patients with biopsy proven NASH and diabetes. Other therapies are not recommended strictly for patients with NAFLD and NASH, but encouraged to treat dyslipidemia such as statins or decrease triglycerides with omega-3 fatty acids. I'd like to point out that obeticolic acid, while approved for PBC, should not be used as of yet for the treatment of NAFLD and NASH. And the Asian Pacific guidelines recommends GLP-1 receptor agonists as there is evidence that it improves the histology of NASH and induces weight loss. There is evidence that metformin, while not effective in the treatment of NASH, not only decreases the risk of hepatocellular carcinoma in a meta-analysis, but also improves survival in patients who develop hepatocellular carcinoma and therefore should be considered in the armamentarium of standard of care approaches to insulin resistance. Statins has also been evaluated in patients with NASH and NAFLD, and in patients, open-label study of patients with biopsy proven NASH received a statin 10 milligrams a day for 52 weeks, improved liver enzymes, and resolved NASH in 19 out of 20 patients. A meta-analysis of studies of the use of statins in patients with NAFLD clearly state that statins are indicated for cardiovascular risk reduction in all our patients, and that statins improve LDL cholesterol and liver function and are safe to be used in patients with NAFLD. However, consistent histologic data to support the use of statins for the indication of NAFLD and NASH are still pending. However, I must add that statins in a meta-analysis have been shown to decrease portal pressure, as well as decrease the risk of hepatocellular carcinoma, suggesting that statin use may in fact improve clinical outcomes in our patients. Shifting a little bit to the role of vitamin E, a meta-analysis of vitamin E in over 1300 patients with NAFLD across 15 randomized controlled studies, understanding that there is a variation across these studies in the definition of NAFLD, and that these studies are moderately small in sample size, suggested that vitamin E, in fact, across all studies, tended to improve steatohepatitis and may have a very marginal, if any, improvement in fibrosis. The most promising patients for vitamin E treatment are obese patients, ages 15 to 50, who have elevation of ALT, with a daily intake of 400 to 800 international units a day, and the liability, ability to lose weight. And there's also evidence in a single center study of patients with biopsy-proven NASH and bridging fibrosis or cirrhosis that use of vitamin E improves transplant-free survival and decreases the risk of hepatic decompensation. Now, in the PIVEN study, which was a 96-week phase 3 trial of 247 patients who were non-diabetic, non-cirrhotic patients with NASH, the study evaluated vitamin E, 800 international units a day, versus pioglitazone, 30 milligrams a day, versus placebo, with the primary outcome being improvement in steatohepatitis without worsening of fibrosis. And as you can see here, vitamin E compared to placebo improved NASH, and pioglitazone approached significance in improving NASH, but was not significant. Neither drug improved hepatic fibrosis. And there are caveats, because the effects noted on NASH were not sustained. Vitamin E may increase mortality due to hemorrhagic stroke and increase risk of prostate cancer. Pioglitazone may increase the risk of weight gain, fluid retention, and increase risk of bladder cancer or osteoporosis. Pioglitazone in patients who are non-diabetic, who are diabetic, has been studied in a systematic review and meta-analysis of 516 patients. And in patients who had NASH with advanced fibrosis at baseline, pioglitazone use appeared to favor an improvement in hepatic fibrosis. And in a study from Ken Kusey out of the University of Florida, which was a randomized placebo-controlled trial of patients who had NASH with prediabetes or diabetes, pioglitazone improved the NASH score, resulted in resolution of steatohepatitis over placebo, but did not improve fibrosis. Shifting to anti-obesity therapies, those that are FDA approved, bariatric endoscopy, and bariatric surgery as treatment options for NASH. The guidelines do make comments about pharmacotherapies for the treatment of NASH. And I specify here GLP-1 receptor agonist as FDA approved weight loss drugs, and also SGLT2 inhibitors, as this class of drugs does induce some weight loss. GLP-1 receptor agonists are approved for the treatment of obesity and diabetes with senaglutide having the most potent weight loss effect and has been demonstrated to improve NASH without worsening of fibrosis. And SGLT2 inhibitors across many different studies has been reported to improve liver triglycerides by magnetic resonance spectroscopy and biomarkers of hepatic steatosis, including even a reduction in liver stiffness. GLP-1 receptor agonists, specifically senaglutide, has been compared in the sustained three, seven, and 10 studies to other GLP-1 receptor agonists. And as you can see, has the most potent weight loss effect compared to exenatide, galaglatide, or even loraglutide. And in non-diabetic obese subjects, when studied in a double-blind randomized phase three trial in adults with overweight or obese states with comorbidities who did not have diabetes, all subjects received lifestyle intervention with counseling, hypocaloric diets, and increased physical activity, and senaglutide, when escalated to a dose of 2.4 milligrams weekly, resulted in a nearly 15% reduction in body weight, with 32% of patients achieving greater than 20% drop in body weight. And how does this apply to NASH? Well, loraglutide, when studied in a 48-week phase two study of 52 subjects at a dose of 1.8 milligrams versus placebo, resulted in NASH resolution without worsening fibrosis in 39 patients who received loraglutide compared to 9% in placebo. And this correlated to a change in weight and a reduction in ALT. And senaglutide has also been studied in a 72-week trial of 320 patients with biopsy-proven NASH and fibrosis stage one to three. And senaglutide was studied at increasing doses up to 0.4 milligrams subcutaneously daily, with the primary outcome being resolution of NASH without worsening of fibrosis. And senaglutide at all the doses studies met this endpoint overall in all randomized patients, as well as those patients with fibrosis stage two or three. The primary outcome, however, did not result in an improvement in fibrosis. The safety of GLP-1s in NASH is consistent with that observed with other trials and disease states. And the most common side effect of GLP-1 receptor agonists in NASH was gastrointestinal. And as you could see here, there was a dose-dependent reduction in body weight and a dose-dependent improvement in hemoglobin A1c. And although senaglutide did not improve fibrosis in this trial, as you can see, those patients who received senaglutide were less likely to have worsening of fibrosis, with only 4.9% of patients who received senaglutide, 0.4 milligrams having worsening of fibrosis compared to 18.8% on placebo, suggesting a potential protective effect. What about SGLT2 inhibitors in NASH? In the e-LIFT randomized open-label study of empegliflozin versus standard diabetes treatments in 42 patients with diabetes in NASH, empegliflozin improved liver fat content and improved ALT compared to placebo. And in the separate study, double-blind placebo-controlled trial of 37 patients with diabetes in NASH, cannafliflozin at a dose of 300 milligrams daily decreased hepatic triglycerides, and this effect was correlated with weight loss. In an umbrella review of seven systematic reviews of SGLT2 inhibitors, including anywhere between 67 to 498 patients, four of these systematic reviews evaluated liver enzymes, four evaluated liver fat, and two reported changes in fibrosis biomarkers. None of these studies were high quality. Five of the systematic reviews indicated that SGLT2 inhibitors decreased liver fat and liver enzymes. One study, a single arm histologic study, showed an improvement in steatosis, but none of these studies showed an improvement in liver fibrosis. Bariatric surgery does improve steatohepatitis in the majority of patients and also improves the metabolic risk factors for NAFLD and NASH. And in a trial out of Lille, France, demonstrated an improvement in hepatic fibrosis over five years with 63% of patients completely resolving mild hep fibrosis and nearly 45% of patients even showing an improvement in advanced hepatic fibrosis. The last statement I'd like to make about endoscopic bariatric approaches, where do they be gastric via aspiration, sleeve gastroplasty, or balloon, or small bowel endoscopic therapy, such as barrier devices, mucosal ablation, or incisionless anastomotic systems, that these approaches have shown a weight loss dependent improvement in the biochemistry with a small intestinal endoscopic bariatric therapy showing an improvement in insulin resistance and weight loss in dependent and independent pathways and even alterations in incretin and gut microbiota. However, large randomized controlled trials are currently lacking. So approach to treatment of NAFLD and NASH is founded on weight loss, irrespective of the approach used to achieve such weight loss and the treatment of diabetes, minimizing cardiovascular risk factors with the armamentarium of therapies that we currently have on hand and targeted pharmacotherapy using vitamin E for non-diabetic pre-cirrhotic patients with NASH, pioglitazone in diabetic pre-cirrhotic patients with NASH. So my takeaway is that in the absence of FDA approved therapies for NASH, utilize available therapies for the primary and secondary benefits, strive for weight loss in a patient-tailored and individualized approach, optimizing all metabolic risk factors. Here are the approaches that improve NAFLD and NASH, and certainly we can get a benefit in clinical outcomes utilizing other therapies such as metformin and statins to improve patient morbidity or mortality overall. Thank you.

Pediatric NAFLD

maternal obesity

childhood NAFLD

cardiovascular outcomes

early detection

intervention

obesity

insulin resistance

glucose intolerance

hepatic steatosis

major adverse cardiovascular events

lifestyle changes

weight loss

exercise

dietary modifications