Good afternoon. Welcome to part three of the postgraduate course. This is session four on alcohol-associated liver disease. My name is Joan Gizopo, and I'm professor of medicine at Harvard Medical School, and my co-moderator is Dr. Jean Im from the Icahn School of Medicine, Mount Sinai. It's our pleasure to introduce this session. The first speaker is going to be Dr. Michael Lucey, who is professor of medicine and chief of gastroenterology and hepatology at the University of Wisconsin School of Medicine and Public Health. Dr. Lucey received his degrees of MB and MD from Trinity College in Dublin. He received training in gastroenterology and hepatology at St. Bartholomew Hospital, London, King's College Hospital in London, and the University of Michigan. His research interests include alcoholic liver disease, viral hepatitis, and liver transplantation. Today, he will tell us about the changing phase of alcohol-associated liver disease. Dr. Lucey. Hello. My name is Michael Lucey. I am the chief of gastroenterology and hepatology at the University of Wisconsin in Madison, Wisconsin. Today, I'm going to talk to you about the changing phase of alcohol-associated liver disease. Here are my disclosures. They're not relevant to the topic I'm discussing. I'd like to first start with a review of public health data on alcohol use and its effects, starting from a global perspective and then going to the United States. And I will also touch on the impact that COVID has had on use here in the United States. I'm going to spend a lot of time referring to deficiencies that exist in recognizing alcohol use disorder and alcohol-related liver disease and the social background to those deficiencies, including stigma. There are also disparities in treatment of alcohol use disorder and alcohol-related liver disease. And one of the reasons for our failure to treat alcohol use disorder in patients with alcohol-related liver disease is the reluctance of gastroenterologists and hepatologists to take on this task. And I'll address that. All of this will lead to a recognition of future needs, where we need to change to try to improve both the recognition of alcohol use disorder and the treatment of alcohol use disorder in order to improve outcomes for patients with alcohol use disorder and alcohol-associated liver disease. Alcohol consumption around the world. In 2016, approximately a third of people were current drinkers, 25% of women and nearly 40% of men. So, it may come as a surprise to you as a listener to this talk that the majority of adults in the world have not had alcohol in the previous year. However, approximately 2.4 billion people globally are current drinkers, 1.5 million men and nearly a billion women. And globally, the mean amount of alcohol consumed was 0.73 standard drinks daily for women and 1.7 standard drinks daily for men. At the base of this slide, you will see the references that you can look up to get further information on what I have mentioned. And this comes from a wonderful paper from the Lancet, which I'm going to show in again in a moment. And it has this figure which shows the different prevalences of alcohol consumption across the world, reflecting different societal attitudes and behaviours. Alcohol is a great source of injury, ill health and death throughout the world. In 2016, 2.8 million deaths were attributed to alcohol use. Just over 2.2 million percent of the total age-standardized deaths among women and nearly 7% of the total age-standardized deaths among total age-standardized deaths among men were attributable to alcohol. And alcohol use was ranked as the seventh leading risk factor for premature death and disability. When we look at a younger population, defined here as less than 50 years of age, alcohol use accounted for nearly 4% of deaths in women and over 12% of deaths in men worldwide. And this is again taken from that Lancet paper I referred to, which found that the global burden of disease attributable to alcohol use was large and greater than had been previously predicted. And that the impact rises with consumption. That is shown in the panel on the right. Indeed, the authors went on to say that the level of consumption that minimizes health loss due to alcohol use is zero. Their conclusion was that public policies regarding alcohol should aim to reduce total population-level consumption. What about alcohol-related liver disease? Well, alcohol-related liver disease is the cause of half of all of the deaths due to liver disease worldwide. And I refer you to this excellent paper in the Journal of Hepatology for further information on this global impact of alcohol on liver disease throughout the world. So, let's turn now to the United States. Between 2011 and 2015, excessive drinking was responsible for an average of nearly 100,000 deaths per annum, 255 per day, with 2.7 million years of potential life lost. That is 29 years lost per death on average each year in the United States. This high number for potential years lost per death is due to the fact that alcohol-related deaths tend to occur in younger people, in contrast to the other great causes of death in society, which are often related to degenerative diseases and to cancer, which have a proportionately greater impact on older persons. The estimate in this data set comprised deaths directly related to alcohol use, but also deaths partially attributed to alcohol, and deaths due to someone else's drinking, an important concept, for example, deaths from motor vehicle accidents where the driver had been drinking, but other persons were killed as well. So, all of those deaths are included in this estimate. Now, with regard to alcohol-associated liver disease, it is the leading chronic cause of death directly attributable to alcohol. So, when you look at this impact of alcohol-related death, the biggest single component is alcohol- related liver disease, with the numbers shown on this slide. And this is a slide from an important paper by Elliot Tapper and Nihar Parikh at the University of Michigan, and you're going to see several papers from this Michigan group in what I'm going to show in a moment. This looked at trends in mortality due to liver disease by age group in the United States. So, this is the changing mortality, and you can see it's increasing across all the age groups. And it's increasing deaths due to alcohol use disorder, but within that, the greatest burden of increase comes from alcohol-related liver disease. And you can see that alcohol-related liver disease and cirrhosis overlap, so that alcohol-related liver disease is the cause of any increase in cirrhosis-related death seen across these years in these age groups. The greatest impact occurs in this younger group, 25 to 34 years of age. This is also a paper from the Michigan group, this time written by Jessica Mellinger, showing that the high burden of alcoholic cirrhosis using a privately insured data cohort, again from the United States, showing over the years that there's increased prevalence of alcohol-related cirrhosis in both men and women, but the increase is greatest in women. So, I mentioned earlier that public policy could try to address this, and here are two panels to give you some information on that topic. The first is from Great Britain, from a paper called Calling Time, and it shows along the horizontal axis the years from 1960 to 2000. And you can see in the heavier black line the real price of alcohol, the price relative to income. So, the true price of alcohol has declined over those years. At the same time, the consumption of alcohol per person over 15 has increased, suggesting a linkage between real price and consumption. The panel on the right looks at alcohol policies using a composite score called the alcohol policy score, which includes many public policies related to alcohol, and it can declare the states as being either liberal about alcohol or stringent. And you can see that there are states with liberal policies and high mortality. This is a valuable paper. It's very recent, as you can see, but I recommend it to you because it also shows that those states that changed their policies to become more stringent appeared to find a reduction in mortality. Can societies try to do this? Well, the one that has got the most attention is Scotland, where they have introduced something called the MPU, the minimal pricing unit. In other words, the cheapest alcohol is the one that's causing the most damage in this assessment. And so, if you bring up the price of that, this is cheap lager, cheap wine, and make the persons who are using their disposable income to purchase this just have access to less of it, you may see a change in amount consumed. And at least in the early data shown here, comparing Scotland to England, the imposition of the minimal pricing unit was followed by a reduction in the grams of alcohol purchased per adult per household, aggregated by week. We need longer data to see whether this change continues and whether it is followed by an impact on alcohol-related mortality and mortality from alcohol-related liver disease. What about COVID? Here are three panels about COVID. The first shows that during COVID, the amount spent on alcohol and the amount spent on tobacco rose. The middle panel shows that during the months when COVID was worst, that's the first phase, essentially, the amount of alcohol consumed rose. And the third is from Mount Sinai, New York, by Stephanie Rutledge, and it's a single-center review of what happened in their hospital. When they closed down wards to accommodate patients with the worst outbreak of COVID, they saw a reduction in admissions for alcohol-associated liver disease. When statewide quarantine processes were put into place, they then saw a gradual increase in the admissions with alcohol-related liver disease to a peak higher than they had seen before. So, you can see more alcohol being purchased, more alcohol being consumed, and then after a brief lag, more alcohol-related acute injury. Now, here is an interesting paradox, and I've called it a dual paradox because it affects both alcohol use disorder and alcohol-associated liver disease. First is that alcohol use disorder tends not to be recognized by healthcare professionals, so that when patients finally are recognized as having alcohol use disorder, they can relate to many prior contacts with the health system where they were not recognized as having an alcohol problem. So, that's a missed opportunity. The second missed opportunity relates to alcohol-related liver disease, which is detected at early stages of injury much more infrequently than for other forms of liver disease, and this is true worldwide, which leads to the rhetorical question, why are we failing to recognize early stages of alcohol use disorder or early stages of alcohol-related liver disease? Well, one reason is that it's about stigma. So, what is stigma? It's a mark of infamy, disgrace, or a reproach, and the term alcoholic is stigmatizing, and our actions and our attitudes can reveal stigma. And to illustrate that, I took this vignette from an old paper by Dr. Groves, a psychiatrist from Harvard, published in 1978, and in it he has this vignette of a 45-year-old man known as Old George, well known to members of the staff of the emergency department. He's been seen there hundreds of times for everything from GI bleeds to a subdural hematoma after a fall, and now he's been admitted with what's described as wildly bleeding esophageal varices. Dr. Groves writes, the staff worked frantically through the night, pumping in whole blood as fast as it would go, but at 4 a.m. the intern pronounced Old George dead, and the junior resident muttered, thank God, under his breath, and the senior resident said, amen, quite audibly. Now, I reread this recently for this talk, and when I did, I was initially struck by thought that maybe this is overly harsh on the junior doctor's present, but then I thought this is correct, because these stigmatizing attitudes start early. They're learned early. They're learned as medical students, and then as interns and residents. We also see significant disparities in the treatment of alcohol use disorder in patients with alcohol-related liver disease. So, the panel on the right relates to liver transplantation and shows that Black patients are less likely to be placed on the transplant waiting list or less likely to be transplanted than white patients, and the panel on the right shows that women are less likely to receive alcohol use disorder treatment when they present with alcohol-associated cirrhosis than their male counterparts. So, we now see that there are failures of recognition and failures of offering treatment. Why are we failing to offer treatment? Well, here are two papers that are worth looking up. I apologize for the ugliness of the slide, but the important messages are here. The first is by Jean Im, and it's a survey conducted by the special interest group in alcohol of the ASLD, the SIG, and any of you who are watching who are interested in alcohol and alcohol-related liver disease, I strongly urge you to join the alcohol SIG of the ASLD. You learn a lot, and as you can see, you can take part in important studies. This study showed that when asked to recall their practice patterns, 60% of providers said that they referred alcohol use disorder patients for behavioral therapy, but three-quarters said they never prescribed alcohol use disorder pharmacotherapy, and the reason was that they were uncomfortable with it. They recognized that they did not have adequate education in addiction medicine, and almost all said they would like more addiction medicine training in their fellowship trainings. This is probably already an overly rosy view of actual practice, because that is shown in the paper below by Rogal et al., which is from the VA. The VA has a well resourced and well-constructed infrastructure for mental health care, and this looked at patients with cirrhosis who were given a new diagnosis of alcohol use disorder, and then what percentage in the following six months were referred for treatment. So, treatment was available. Was it made use of? And only 14% received treatment, and it was almost exclusively behavioral. Almost nobody got pharmacotherapy, so the 60% in the upper panel probably is an overestimation of what we actually are doing in practice. So, that's the bulk of my talk here. I started by showing the public health impact of alcohol use, both globally and nationally, and it's severe, and then it has been made worse by COVID, both on purchase of alcohol and treatment of AUD in ALD patients. And there's a more broad-based problem here that we are failing to recognize. Alcohol use disorder, that includes at the level of primary care as well as subspecialty care, and we're failing to treat alcohol use disorder in patients with alcohol-related liver disease. Why is that? Well, it's obviously many reasons. One is stigma, but there are also deep-seated disparities in the way we treat patients, and these are racial disparities and gender disparities. I think if we look for data, we'd also find rural and urban disparities. We'd also find potentially age-related disparities and disparities related to means and poverty. And then there's a problem within the profession itself. As gastroenterologists and hepatologists, as nurse practitioners and physician assistants working in liver disease, we are reluctant to treat alcohol use disorder, and this reluctance stems from a lack of training, a lack of comfort in this area, and that leads to future needs. We really need a new culture regarding teaching and practice, regarding alcohol in education of doctors, advanced practice providers, nurses, and the other healthcare professionals involved in the care of patients with alcohol-related liver disease. So that completes my talk. Thank you very much. I hope you found it valuable, and it prepares you for the next talks to come. Thank you very much. It's my great pleasure to introduce Dr. Ramon Batalhaer from the University of Pittsburgh, where he is Chief of Hepatology of the University of Pittsburgh. He received his MD from the University of Valencia School of Medicine in Spain. He trained in hepatology at the liver unit in the hospital clinic of Barcelona, and his PhD is from the University of Barcelona School of Medicine. Dr. Batalhaer's main research interests include liver fibrosis and the mechanisms, determinants of mortality, global public health aspects of alcoholic liver disease. He's also a major scientist in translational research in alcoholic liver disease. Today, he's going to talk about giving your patient the best chance of sobriety. Hello, my name is Ramon Batalhaer. I'm from the University of Pittsburgh, and it's for me an honor to talk to you in this postgraduate course. I would like to thank both Laurie and Mary for this kind invitation. And my task today is to discuss with you and to try to guide you how to keep our patients with alcoholic liver disease sober for a long time, ideally permanently, so they can improve the liver disease. So I have no disclosures for this presentation. This is a little bit the summary of what I will cover, but I have to say a disclaimer here. There is not much evidence in this field. There is a single randomized trial how to treat alcohol use disorder in these patients. So besides some evidence, I will also talk as an expert because I have been 20 years devoted to these patients, and I will share with you some of my clinical experience. We'll discuss the impact of alcohol consumption in patients with ALD. How can we identify who is a drinker because we do not identify a patient with drinking? How can we cure it? Also, how can we apply personalized medicine in this field and which is the role of pharmacological therapy and also motivational? And finally, how in a hepatologist within a multidisciplinary team can also help these patients within a team how to stay sober. Let's start with the natural history of ALD that I think is known by all of you. That alcohol consumption will lead to steatosis, fibrosis, cirrhosis, ATC, and eventually decompensation and death. The problem of this field compared to anaphylactic is that we mostly see these spaces where they decompensate, when they become jaundiced, when we go too late to this field and we need to move this field at the earlier stages, and we need to do campaigns for early detection of silent fibrosis, even silent cirrhosis in this field. So most of the studies have done in decompensated disease and we need to start doing early detection in these patients. Well, there are three or four studies showing that abstinence can impact the natural history of survival of these patients. This is a study led by Caroline Lackner in Austria. We showed that patients with alcoholic liver disease in a compensated state, when they have F3, F4, they have at 10 years, 40% mortality, which is a lot, much more than NAFLD, much more than other type of liver disease. This is a strong argument to detect silent cirrhosis in an urgent way in hepatology. And this is the impact of abstinence. Abstinence impacts survival, but some patients, despite being abstinent, still can die. So we need to detect these patients at earlier stages. Here is a study that we did in the past with Jose Altamirano and the team in Barcelona. We follow up patients with alcoholic hepatitis that survived the index episode for 15 years. Initially, we hypothesized that controlled drinking was better than heavy drinking, but the results did not support this hypothesis. Any kind of drinking impairs survival of these patients. I learned from this study several things. Number one, many patients that remain abstinent still die. So we need to offer transplant from patients. And again, and I will repeat this concept many times with my presentation, we need to go earlier in these patients. Second thing that I learned, a lot of patients relapse, more than half of them. And some of them relapse very later, very late in life. So what that taught me that I never forget about talking about alcohol problems with my patients. In every single visit, I'll always bring and discuss the topic of alcohol. Even patients sober for three, four years, they still can relapse later on. And finally, we found that young people, especially women, are more prone to relapse. So be careful because these young people that initially do very well, they're very scared, they yellow. In the long-term, they have high risk of relapse and we have to be really, really on top of that. The third piece of evidence was a recent study that by Sherry Rogal in the VA here in Pittsburgh, showing that some good and bad news. The good news is that treating alcohol use disorder in cirrhotics improves the mortality at 180 days and long-term, but also prevents readmission and decompensation. But the bad news is that we only were giving these to 14% of the patients. We only treat 14% of the cases, the alcohol use disorder. To make an analogy, can you imagine that we will only treat 14% of the cases in hepatitis C and hepatitis B? That is unthinkable. So we need to be much more proactive treating alcoholism and alcohol use disorders in our liver patients. The next thing I want to discuss is how we identify that the patient has alcohol as a pattern or as an underlying cause of the liver disease. That is more challenges that we may think. Many patients are labeled with naphalete or they have also hepatitis C and they under-report the alcohol use. So the first thing, and I don't have time to go into detail to all these parameters to identify alcohols to use is that sometimes the physical exam can reveal alcoholism. For example, look at these Dupuytren or the renal FEMA. Some of the patients that I have seen in my clinic or when I run, that they are labeled as naphalete, they have these signs and you have to suspect that alcohol is also behind. But also they can have multisystemic involvement. For example, they have peripheral neuropathy in the absence of diabetes. They should tell us something. Or other patients have multiple fractures or also they have social isolation, et cetera. They are a little more prone to have alcohol. So we have to have better clinical eye in suspecting alcohol as a cause of the liver disease, even in patients that have been labeled with other causes of liver disease. Well, this slide is, I don't have time to go into detail to this. This is based more in clinical experience. How can we reveal alcohol use under-reporting among patients with liver disease? There are no studies showing the impact or the prevalence of other under-reporting or how can they, under-reporting can impact for the relapse, for example. I have to say that we have to build some trust with the patient. We need some time to ask this. We can ask that in one or two minutes. We have to be sensitive to the stigma. Gender is the biggest stigma. Religious can be a stigma. The background of the patient is important. We need to be culture sensitive sometimes. We obviously can ask family members, firstly, relative members. We have to take a look at them. Sometimes just the body language can tell us if the patient is drinking or not. There are several techniques that I learned from the counselors that were running with me and visiting patients with me, that the overshooting technique. I tried to tell, what is the last drink you had last night or two days ago? And they tell me, no, it's one week ago. So they feel entitled to tell me that one week ago is a very acceptable time for me for the drinking. So there are other techniques. And obviously besides asking their relatives, besides asking the patients, we need to have objective measures. And there is a big role for biomarkers. And I would like to discuss a little of the biomarkers with you. This is a figure that we included in a review article that we did with Joaquin Cabezas and Michael Luzzi a couple of years ago about potential biomarkers. Obviously, one is GGT. GAT, sometimes we don't use it enough. And when I see enough of the patients with it, GGT of 300, 400, I have a red flag. They know when they present patients for transplant company in my hospital, then they need to stop in the GGT because I always take a look at the GGT. AST-ALT ratio is also a suggestive of alcohol, but be careful that patients with cirrhosis and malnourished patient can have an AST more than ALT with alcohol, be careful. What about the biomarkers? In the blood, people use a lot of blood biomarkers. Alcohol, after 20, 24, 28 hours disappears from the blood. Blood alcohol is for the intoxication, but I don't use blood alcohol for any diagnosis of alcohol use disorder. In the urine, and this is, sorry, I'll pause this. This ETG, ETS in the urine, they are good biomarkers. They reflect three to four days of alcohol, but be careful. The center has to be equipped for addiction testing. If there is no nurse that goes with the patient, make sure that the patient doesn't bring a container or the patient goes with someone else to the toilet, they can change, and savvy alcoholists, they can change the urine. We're using a lot of PATH, and PATH is changing our life in the clinical practice, at least here in Pittsburgh. They have two advantages to me. Number one is blood-based, so the patients cannot change the urine, et cetera. Number second, it captures three to four weeks of alcohol, so patients really, even if they stop drinking a few days before the visit, you capture them. And the third thing is a very quantitative thing, and I will show you here a tweet that I sent a few weeks ago that said the same. I use Serum PATH to monitor ALD patients. It's reliable, captures three to four weeks of consumption, and when I see levels more than three or 400, this suggests that the patient may need to be admitted to detox. So really, I have seen up to 2,000, 1,000, 1,500, all these patients need detox. There's no way they will stop drinking on their own. This is very important, and I think we need more studies on PATH, not only to diagnose ALD, but also to diagnose NAFLD. In my opinion, the N on NAFLD needs a PATH, confirming that the patient is not drinking, and I am waiting for the studies checking this assumption. Okay, how to do a personalized medicine to manage ALD? Look, when we talk about personalized medicine, everybody thinks about genes, the genetic polymorphism, et cetera, but personalized is much more than the genes. In alcohol, look at these common associated conditions. I know we don't have time to ask for that. I know hepatologists are not well-trained to ask for that, including myself, but if you don't detect the underlying cause that led this patient to drink, and the patient is coping any of these causes with the drinking, it's very difficult to treat and to beat the alcoholism. PTSD in young women is very common. Sexual abuse, we have seen a lot of young ladies, and we need our counselor to be trained in that regard. Depression, anxiety, sleep disorders, pain, and many others. We need to identify patients with severe psychiatric conditions and convince them to go to a motivational interview and to a psychiatrist. We need a multidisciplinary team to beat the disease. We alone are unable to treat the cause. The antiviral therapy for this disease, the main technology, is to treat the alcoholism, and we need help, and we need to be better trained to do that efficiently. Well, there are a few studies in different type of motivational intervention in these patients. This is a recent study published by our friends for the Mayo Clinic that patients surviving an index episode of alcoholic hepatitis were sent very early to an addiction specialist, and they found, both in a test cohort and a validation cohort, that that early referral saved lives, and they can stop drinking and have a lower mortality and less readmissions. At least in my clinical practice, every, especially new patient, these young new generation of patients, and in America, we're seeing a lot, especially women with alcoholic hepatitis, we need to see them very early now with our patient clinic. We cannot send them two or three months after an episode. We need, ideally, to say the next available week, the next week or two weeks, we see these patients, and at least in my clinic, it works very well to have the addiction counselor in our clinic physically. The patients are very unlikely to go to another clinic another day with problems with transportation, with money to pay for the visits, et cetera, so I really encourage you to have the team physically together in the same place. That helps a lot, in my experience. What is the role of pharmacological therapy? We have a review article that is in press and will be released very shortly that was done by Juan Pablo Arap and also B.S. Zagg and Leggio from the NIH, Serpaena and Ms. Eze. We review a little of the current therapy. I have to say that it's only a trial, and the FDA approved for a UD doesn't mean that they are approved for alcohol liver disease patients. This is an important message. So, disulfiram, be careful. I have seen four patients dying in my life because fulminant hepatitis due to disulfiram in cirrhotics that were underdiagnosed and they were given disulfiram. So, disulfiram can only be given to patients with a significant fibrosis. Be careful. It's a potential hepatotoxic. Acamprosate, it seems effective. It seems safe in my experience. I have not seen hepatotoxicity, but a question is that they're not clinical trials assessing acamprosate, randomized, double-blind, et cetera, in patients with ALD. It seems safe. I don't get it when patients have more than 5, 6, 7, 8 bilirubin. Wait a little bit until the patient's liver function has improved, but it could be useful. And in the absence of clinical trials, at least the expert opinion tells you I have not seen major causes of hepatotoxicity. Naltrexone, which another FDA approved, is also potentially hepatotoxic, much less than disulfiram. I wouldn't give it to patients with any degree of liver failure. I will wait until the liver function recovers minimally. But it's a very strong and it's very effective as a therapy AUD, and I also wait for clinical trials seeing not only the efficacy, but also the safety of this naltrexone in patients with advanced liver disease. And finally, Baclofen, which is a GABA agonist, that is not FDA approved yet for alcohol use disorder, but is the most serious study. It was done years ago by Adolorato from Italy, and it was a study done in compensated alcoholic cirrhosis. You see the people treating with Baclofen has less chances to resume the alcohol intake. It's basically indicated the patients with craving for alcohol. It seems very safe. We start 510BID or TID sometimes can give some dizziness. So sometimes I start at night when the patients are prone to dizziness, but it's very safe in other side effects. And I give it up to 20 milligrams TID, 80 milligrams is the max dose that you can give. But one advantage is that there are few reports showing that Baclofen can be good for withdrawal syndrome, or even binge eating, which is a problem. Many of these patients shift alcohol to sweets, and they gain a lot of weight, but even for smoking cessation. So for any oral anxiety, it seems that Baclofen can work, but as I say, it's not FDA approved yet, but it's the most serious study and the best public design study in the alcohol liver disease was done with this Baclofen. And finally, the last couple, three minutes of my talk is how the hepatologists can be useful within a multidisciplinary team to treat alcohol disorder. So we need to identify the alcohol is a problem. We are not good. We don't have a good clinical eye suspecting alcohol. We will fail. We cannot treat something that we have not identified, obviously. We need to be able to work in groups, and that, believe me, is more difficult than we think. We need to have transversal groups with addiction therapies, with psychiatrists, with neurologists, with hospital workers. We have to be used to work in teams, not even being the boss. We are part of a team around a patient, and the patient is in the middle. The patient is the center. Finally, we need to be better trained, and I will show you a survey showing that we even recognize that we're not well trained for motivational, also for drug therapy. We need to identify, as I said before, the underlying condition. We need to devote a little time to ask for that and be trained to be sensitive how to ask these questions in an effective way. And finally, we need to convince the patient to go to the addiction therapist, and we need to be persuasive, and we need to be motivational, and I have learned that over the years. I was not very efficient initially. I have learned a lot, and my main teaching point has been to work with other specialties together with me. All the addiction therapies have taught me a lot, and most of the things I know is thanks to them, so I invite you to work together physically also besides other specialties, and this is a very interesting and very important survey that was conducted by the ASLD SIG4LD. Jeanine was the first author. Basically, they did several surveys to see which was the drug more used. Baclofen was the number one used drug, but the graph that caught my attention is what is your barriers to give pharmacotherapy to treat a UD in these patients, and the number one barrier was lack of training and unfamiliarity, so ourselves, we are now evidence-based showing that we need training, and we're working in that to provide some material for training. I know NHAAA has in the webpage good training material as well, where we need to be better trained how to motivate these patients and also how to use the drugs in a safe and effective way, and the last slide on my talk is a picture that I took from the human castles, we call in Catalonia and Barcelona, there are many layers of people, there are nine layers of people, it is all the hands, that the guy going to the second layer needs to be able to be stable in the second layer. So this is a very good analogy for our patients. We need a lot of hands to treat our patients. We cannot work anymore by our own, we need help. And we need to work in a multidisciplinary team. Thank you for your attention. The final talk in our session is gonna be by Dr. Patrick Kamath. It's my real pleasure to introduce Dr. Kamath, who is professor of medicine at Mayo Clinic. He graduated medical school from the St. John's Medical College in Bangalore, and did his internal medicine residency and gastroenterology fellowship training at the Postgraduate Institute of Medical Education and Research in Chandigarh, India. His research focuses on the acute and chronic liver failure, non-alcoholic fatty liver disease, polycystic liver disease, butchiari syndrome, and heterodotoric hemorrhagic telangiectasia. Dr. Kamath also studies alcoholic hepatitis, cirrhosis, and the complications of these diseases, including portal hypertension, variceal hemorrhage, ascites, and hepatic encephalopathy. He is gonna talk about keeping the patient alive with decompensated alcohol liver disease. And I'd like to welcome Dr. Kamath. My thanks to the course directors for inviting me. These are my disclosures. To discuss decompensated alcohol-associated liver disease, let's start with the case scenario. A 38-year-old male with an alcohol use disorder admitted to the ICU five-day history of fever. He's jaundiced, has ascites, and is confused. Total bilirubin is 17 milligrams per deciliter, AST greater than ALT, melasco 36, creatinine 2.8 milligrams per deciliter. The chest X-ray shows pulmonary infiltrates, hypoxemic, hypotension, tachypnea, and leukocytosis, an inflammatory response. The patient has started on piperacillin-tazobactam. Sputum cultures grow klebsiella. But after 24 hours, the patient is more confused and drowsy. Endotracheal intubation is carried out. Serum creatinine rises to 4.8 milligrams per deciliter. The chest X-ray shows new infiltrates, and antifungal therapy is started. So the question for us today is, how do you keep this patient with decompensated ALD and at high risk of mortality alive? To do this, we determine the risk of dying, reverse the alcohol-associated hepatitis if possible, treat the infection, support the failing organs. I will also discuss liver regenerative and bioartificial therapies, and finally consider liver transplantation. First is determining the risk of dying. Which prognostic score should we use? Should it be MELD, the CLIF-SOPA, APACHE, NAC-SELD, the ASIAN score? And I will submit to you that none of these are particularly helpful. Because the C-statistics for these have not been considered, the C-statistics for these have not been consistently shown to be greater than 0.8, which is a threshold for excellent model. And so we are talking about the patient who is critically ill. There have been limited independent validation studies using these scores. And the accuracy of these scores is highest when it's too late to reverse the disease. So let me give you an example. This is the London Olympics, 2012. Usain Bolt, very close to the finish line. So here, we are not going to be saying, Usain Bolt, I predict he's going to win the gold. You're going to be saying, Usain Bolt is winning the gold. So very close to the finish line or very close to death, models are reflective of the dying process and they're not predictive. And so organ failure scores, especially with increasing organ failures, they're describing the dying process and they're not predictive of death. So they are useful when it's really too late to reverse the course of the disease. Alex Louvet from France has combined the MEL-SCO and LEL-SCO for optimal prognostic scoring in patients. So the MEL-SCO at baseline and the LEL-SCO at seven days is useful in predicting mortality. For instance, at baseline, this patient has a MEL-SCO of 21 and on day seven, the LEL-SCO is 0.45. Using this nomogram, the two-month mortality is 15% and the six-month mortality is 24%. So in general, the prognostic factors in these patients are severity of the alcohol-associated hepatitis at baseline and the response to treatment at day seven predicts whether patients are going to be alive at six months. If there's no alcohol relapse and there was an initial response to treatment, such patients will do well in the long run. If there's no alcohol relapse, the patient had not responded to treatment initially, this patient will have a substantial risk of death and liver transplant has to be considered. If there has been an alcohol relapse, then this patient has a high risk of death and in such patients, consider palliative care. So in the short-term period, if you want to keep the patient alive, target liver injury. For the long-term, we have to target alcohol behavior. In general, when a patient is hospitalized and is in the intensive care unit with severe alcohol-associated hepatitis, I use the rule of 50%. This patient has a 50% chance of dying in hospital. If the patient is dismissed, the patient has a 50% chance of readmission in six months. And if the patient is listed for liver transplantation, a 50% chance of delisting. The next, but going on at the same time, is trying to reverse the alcohol-associated hepatitis. Now, the normal hepatocyte takes up ammonia and releases urea into the circulation. The hepatocyte also divides into daughter cells, which is the regenerative capacity of the liver. Alcohol damages the hepatocyte and so that ammonia and other waste products cannot be taken up by the liver and escape into the circulation. Inflammatory cells are recruited from the extracellular matrix. And these inflammatory cells release pro-inflammatory cytokines, which again escape into the circulation. And these inflammatory cytokines also inhibit mitosis. So the end result of hepatocyte injury is waste accumulation, systemic inflammation, and impaired hepatic regeneration. So the course of these patients is they have a systemic inflammatory response because of hepatocyte injury. This is initially a sterile response, but then there's an infection-related inflammation, which worsens the inflammatory response. Patients then progress to organ failure and in the absence of liver transplantation, a reversal of the cause will die. The organ dysfunction, which follows the inflammatory response, includes hepatic encephalopathy and occasionally cerebral edema, high output cardiac failure, acute lung injury, loss of metabolic liver function, and these patients also get cholestatic, adrenal dysfunction, acute kidney injury, and bone marrow suppression, which leads to immunoparesis and contributes to the high risk of sepsis. So what are the potential therapeutic targets in these patients? What can we do to try and keep this patient alive? Patient initially had compensated liver disease, falling alcohol use, developed alcohol-associated hepatitis, was infected, and developed multi-organ failure. So perhaps if we treat the alcohol-associated hepatitis and the infection, support the failing organs, and consider liver transplantation, this patient can be brought back to a stage of compensated liver disease. So how do we go about that? So the initial evaluation in patients suspected to have alcohol-associated liver disease, and that is based on the clinical presentation of heavy alcohol use and recent onset jaundice. Laboratory picture for rapidly rising bilirubin with AST greater than ALT. We want to rule out other causes of jaundice, and these include biliary obstruction, drug-induced liver injury, viral hepatitis, autoimmune hepatitis, and ischemic hepatitis. Once these conditions have been ruled out, and we've diagnosed alcohol-associated liver disease, we treat the alcohol use disorder and the liver-related complications, which are hepatic encephalopathy, gastrointestinal bleeding, infection, and acute kidney injury. And I'll focus a little more in this talk on infection and acute kidney injury. So the stepwise approach to management is if the patient has a typical clinical presentation and laboratory tests, there's a definite alcohol intake history, and there's no potential heptotoxic substance in the last three months. If all these are yes, the patient can be diagnosed to have probable alcohol-associated hepatitis. If a liver biopsy is carried out, and histological confirmation is obtained, this patient has definite alcohol-associated hepatitis. On the other hand, if any of these three is uncertain, the patient denies history of alcohol use, or may have taken a recent medication, then a liver biopsy is required before we can diagnose alcohol-associated hepatitis. If no liver biopsy is carried out, this is only possible alcohol-associated hepatitis. For probable and definite alcohol-associated hepatitis, and a MADRI-DF of greater than 32, or MELD of greater than 20, steroids are started in the absence of contraindications. Leads go checked at seven days, less than 0.45, continue pregnancy loan for four weeks. On the other end, if it's greater than 0.45, consider early transplantation, stop pregnancy loan, consider clinical trials in these patients, and very important, a family discussion on goals of care. The infection also needs to be treated aggressively. So in the intensive care unit, or wherever the patient is, a paracentesis, culture blood, a situs urine, get a chest X-ray, and check lactate. Treat the infection. If you suspect multidrug resistance, use vancomycin and meropenem. No response within 48 hours, consider antifungal treatment. The general measures include fluid resuscitation within three hours, a therapeutic paracentesis to avoid the abdominal compartment syndrome, aspiration precautions, deep vein thrombosis prophylaxis, and also stress ulcer prophylaxis. Blood pressure, less than 60 millimeters of mercury mean arterial. This is septic shock, use norepinephrine. Despite norepinephrine, mean arterial pressure less than 60 millimeters of mercury, this patient probably has adrenal dysfunction, consider hydrocortisone. Support the failing organs, and here I will discuss only acute kidney injury and renal replacement therapy. So this is the approach to the patient with acute kidney injury and decompensated acute liver disease. If there's an acute increase that is within 48 hours and the serum creatinine to greater than 0.3 milligrams per deciliter increase, but the increase is less than twofold and the serum creatinine is less than 1.5 milligrams per deciliter, this is stage Ia AKI. In this patient, risk factor management and monitor for 48 hours. And risk factor management is all nephrotoxic drugs, beta blockers, withdraw diuretics, treat the infection, and plasma volume expansion as required. Renal function improves, monitor the patient. However, if there's no resolution, check again whether all these have been taken care of. And if the creatinine is greater than 1.5, consider albumin, one gram per kilogram for two days. Despite albumin challenge, or if with albumin challenge there's resolution, monitor the patient. Despite albumin challenge, there is no resolution, check urinary biomarkers. If the urinary biomarkers are consistent for the acute tubular necrosis, renal replacement therapy on an individualized basis. Urinary biomarkers are consistent with hepatorenal syndrome, risk-active therapy, and albumin. But often urinary biomarkers are not available or non-diagnostic. In such patients, treat them initially as hepatorenal syndrome with vasoconstrictor therapy and albumin. And consider renal replacement therapy again on an individualized basis. How do we go about renal replacement therapy in patients with AKI and decompensated ALD? If this is acute tubular necrosis, determine whether the patient is a transplant candidate or not. If this is a transplant candidate, dialyze until transplant, and some of these patients might require combined kidney and liver transplant. If they're not transplant candidates, consider dialysis, but more limited. If this hepatorenal syndrome type of acute kidney injury, a transplant candidate, dialyze as bridge to transplant. Not a transplant candidate, no to dialysis. In the intermediate group where the diagnosis is unclear, again, determine whether the patient is a candidate for transplant or not. If the patient is a transplant candidate, a limited trial of dialysis, not a transplant candidate, give the patient the benefit of the doubt, but a shorter trial of dialysis. So to summarize, the decision to dialyze is based on etiology of AKI and transplant candidacy. What about liver regenerative and bioartificial therapies? Unfortunately, not encouraging. So GCSF has been tried for alcohol-associated hepatitis. The meta-analysis of the seven studies on use of GCSF in patients with alcohol-associated hepatitis has shown a favorable effect of GCSF only in Asian studies. So the utility of GCSF in severe alcohol-associated hepatitis needs validation outside of Asia. Liver support systems, again, not very encouraging. There are heterogeneous study groups, but they're largely alcohol-associated liver disease. There are limited randomized studies, and multiple devices have been used for a total of 16 trials. In these trials, the role of liver support systems has not been fully demonstrated, but of these, plasma exchange seems to have had the most promise. And then liver transplantation, which would be discussed by my former fellow and colleague, Michael Charlton, whom I remain very proud of. So the key takeaways on keeping the patient with decompensated ALD alive. Decompensated ALD patients have a 50% risk of dying in hospital, and if dismissed, a 50% risk of free admissions in six months. Meld and Leal scores are used for prognosis. Early intervention is required to save lives, and this is reversing the alcohol-associated hepatitis, treating infection, and supporting failing organs. The benefit of regenerative therapies and liver support devices has not been demonstrated, and consider liver transplantation in selective patients. Thank you. Welcome to session five, where we will be discussing emerging therapy for alcohol-associated liver disease and non-alcoholic fatty liver. My name is Haripriya Madur, and I will be co-moderating the session with Dr. Nora Thoreau. We will begin by a discussion on liver transplantation of alcohol-associated hepatitis, presented by Dr. Michael Charlton from the University of Chicago Medicine. He will be discussing evolving trends, ethics, and outcomes in liver transplantation. This will be followed by a discussion by Dr. Vijay Shah from the Mayo Clinic, discussing emerging therapy for alcohol-associated liver disease. Not only the physiology of alcohol-associated hepatitis, but what are the most promising therapeutic targets. Dr. Stavros Anthikos from Cincinnati Children's Hospital will then be discussing bariatrics and adolescent obesity. And then we will have a fantastic lecture by Dr. Steven Harrison from the University of Oxford on emerging therapies for NASH. Thank you. Hello, my name is Michael Charlton. I'm a professor of medicine at the University of Chicago, where I serve as medical director for the Transplant Institute and director for the Center for Liver Diseases. I'd like to thank ASOD and also the course directors, Dr. DeLeve and Dr. Rinella for the opportunity to speak today about one of the most important and rapidly evolving areas in hepatology, which is liver transplantation for alcoholic hepatitis. The very first adult to undergo liver transplantation nearly 60 years ago was a 48-year-old man with what was described as the next cirrhosis, complicated by hepatocellular carcinoma involving all four liver segments and attachment to the right hemidiaphragm. The MELD score can be derived from the biochemical values and hematologic values in the original description. It would have been 13. The patient had been febrile for two weeks preoperatively. The patient died on post-operative day 22 of events unrelated to his original liver disease. The first series of patients undergoing liver transplantation globally, including at the University of Colorado, expired in relatively short order following liver transplantation. This resulted in the global moratorium on liver transplantation until 1967. It is safe to say, however, that liver transplantation for alcoholic liver disease has been vexing from the very beginning of this procedure. In the 1970s and 80s, the success of liver transplantation increased and became more consistent as hurdles in immunosuppression, the operative technique, and other aspects of this procedure were more or less ironed out. The National Institutes of Health convened a consensus conference in 1983, published in 1984, and there was a subsequent position paper by Drs. Starzl, Dimitris, and Van Thiel published in the New England Journal of Medicine. There were two important points made in these papers relevant to today's presentation. The first is that the conceptual appeal of liver transplantation is so great that the procedure may come to mind as a last resort for virtually every patient with lethal hepatic disease. And secondly, that only a small proportion of alcoholic patients with liver disease would be expected to meet these rigorous criteria. Subsequently, liver transplantation for alcoholic hepatitis was performed in less than half of 1%, 0.5% of all liver transplant recipients between the 1980s and 2010s, as it was generally deemed to be an inappropriate indication. In 2011, there was a seminal publication first authored by Philippe Maturin describing liver transplantation in patients with alcoholic hepatitis. Now, the selection criteria for patients in this paper are important. They identified as potential recipients patients who had a Madry's discriminant function of greater than or equal to 32, a threshold for initiating leukocorticoids, had to have had a non-response to medical therapy, and this was defined as allele score of at least 0.45 after seven days of medical therapy, or a persistent increase in MELD score. Alcoholic hepatitis had to be the first liberty compensating event. Now, there had to be a presence of close supportive family members and absence of severe coexisting psychiatric disorders. And finally, an agreement with support from family members to adhere to lifelong total abstinence from alcohol. This slide shows the Kaplan-Meier survival curve for patients who underwent liver transplantation compared to controls matched for degree of illness, et cetera. And the survival benefits at six months was profound, 23% in the matched controls versus 77% in patients undergoing liver transplantation, a 54% survival benefits associated with liver transplantation. And it's worth pointing out that the six month survival in this publication was considerably less than is obtained currently, which should be in the order of 90 to the low 90s percent. The audience will be familiar with the paper that I just discussed. And I'd now like to pivot towards emerging issues on the frequency of liver transplantation for alcoholic hepatitis. Among the important things to consider about liver transplantation for alcoholic hepatitis is the name or the term itself. Cirrhosis is near ubiquitous in patients who are thought to have alcoholic hepatitis. It has been found in between 90 to 96% of patients who undergo liver transplantation for alcoholic hepatitis. And histological features of alcoholic stator hepatitis itself are only present in between 50 and 59% of explants in prospective studies. And the outcomes are similar regardless of whether cirrhosis is present in the explant. A more accurate terminology is alcohol-related acute liver failure. And this is the term I will use for the remainder of this presentation. The impact of the Franco-Belgian Consortium paper was dramatic. Prior to its publication, the number of centers in the United States performing liver transplantation for alcohol-related acute liver failure was between two and five per year. Subsequent to the publication of the paper by Maturin et al, the number increased in a curvy linear fashion from 12 to the more recent data of 73 centers in the United States performing liver transplantation for this indication. Perhaps unhelpfully, French exports of wine and spirits to the United States have increased from $2.5 billion per annum to $4.5 billion since 2011. Earlier this year, we published a paper looking at the national experience in the United States using the Scientific Registry for transplant recipients of liver transplantation for alcohol-related acute liver failure. This slide shows the percent change or increase in liver transplantation for alcohol-related acute liver failure, over 200% increase in the five-year period of the study, far exceeding that of the 2010 study. We can compare that for NASH, for example. Hepatitis C declined by 70% in this study period. There were 435 patients in the data set that undergone liver transplantation for alcohol-related acute liver failure and about 40,000 patients in the study period overall. Importantly, this now accounts for about 2% of liver transplantation in 2019. The rate of liver transplantation for grass survival was seen to be 92% and 82% for alcohol-related acute liver failure. This was numerically superior or higher than any other indication in the period of study, and there was no trend for attenuated grass survival compared to any other indication. We observed a marked geographic variation in liver transplantation for alcohol-related acute liver failure, according to the 11 UNOS regions. And you'll see that the highest frequency, over 4%, was observed in New England, which would include, for example, Maine, and the lowest frequency in Region 8 that includes, among other states, Iowa. And it could be presumed that this variation in frequency was on the basis of variation in excessive alcohol consumption or need for liver transplantation for this indication. This turns out to be totally untrue. This is a slide of a geographic variation in rates of excessive alcohol consumption. And you will see that Maine has among the lowest frequency of excessive alcohol consumption in the United States, and Iowa has among the highest frequency of excessive alcohol consumption. So the performance of alcohol-related acute liver failure, liver transplantation is totally dissociated with the burden of disease. This slide shows a graphic describing a recent surge in liver transplantation, shown with the dark green line, and candidates with alcoholic hepatitis added to the wait list nationally since before the pandemic in 2018 through to the current period. And rates have increased by over 400%, a fourfold increase in liver transplantation for alcoholic hepatitis in this period of time. The first state-level COVID-19 shelter-in-place order occurred just before this surge occurred, and the authors presume that these two events may have been connected, this increase from 2% to 7% of liver transplantation being carried out for what was described as alcoholic hepatitis. So why the dramatic increase? Is it on the basis of more disease? Was it greater acceptance of this as an indication? Or is it something else? Perhaps the best indicator of the health impact of the well-documented increase in alcohol consumption during the pandemic is on the number of GI, liver, and pancreas-related emergency room visits per 100,000 of the population. And you see that prior to the pandemic, the number of visits was in the range of 450, and since the shelter-in-place order, up to 575. Certainly noticeable increase, but nowhere near the increase seen of 400% for liver transplantation for alcohol-related acute liver failure. So unquestionably, the first state-level COVID-19 shelter-in-place orders coincided with this onset of the steep increase in liver transplantation for alcohol-related acute liver failure. But perhaps more importantly, the UNOS, or United Network for Organ Sharing, liver allocation policy changed at essentially exactly the same time. This slide shows the 11 UNOS regions from which recipients would have been able to attract organ offers from donors under the SHARE-35 system in place prior to the acuity circle. So look, for example, at a donor originating in Chicago. Donor organs would have been offered to potential recipients within the states outlined in this slide. In contrast, acuity circles are agnostic to UNOS regions. So a donor in Chicago would be made available to a potential recipient within up to a 500-mile radius from the geographic location of the donor. So conversely, a recipient in Chicago would be able to potentially compete for a donor organ from donors anywhere within the 500-mile radius. The potential impacts of this change in allocation system is hinted at in this slide, which is a homunculus in which the size of the peaks is proportional to the population density of different areas, particularly metropolitan areas across the United States. And you'll see that a recipient, potential recipient in Chicago will now be able to draw organ offers from a population base that is millions greater than was possible under the previous allocation system. And this is important as in our analysis of national trends in liver transplantation for alcohol-related acute liver failure, we found that patients with alcohol-related acute liver failure have about half of the waitlist mortality when compared to any other indication for liver transplantation. So the ability for a patient with alcohol-related acute liver failure to await and survive for a suitable donor organ offer is much greater than for other indications. So the MELD-based allocation systems clearly favor alcohol-related acute liver failure. And the UNOS concentric circles policy has clearly exacerbated the advantage for alcohol-related acute liver failure. So let's turn to other aspects of liver transplantation for alcohol-related acute liver failure. So based on what we've learned from the Franco-Belgian consortium study, if you have 50 patients shown here with severe acute alcoholic hepatitis, there will be about a 60% response rate to medical therapy, whether it's corticosteroids or general supportive care. Among responders to medical therapy, there's a 25% mortality. Among non-responders to medical therapy, 40% of patients with severe acute alcoholic hepatitis, there's a 75% mortality. And among transplanted patients, about a 10 to 25% mortality, but there were 80% of patients not eligible for liver transplantation. So going forward, the most pressing question facing us is how many, who among the ineligible non-responders, 30% of severe alcoholic hepatitis patients, might also be appropriate transplant candidates. This slide shows reasons for ineligibility from the Matterin study. And by far the most common reason for ineligibility was addiction or social factors. This is potentially modifiable in terms of the interpretation for eligibility. Comorbidities more or less fixed in being strong relative or absolute contraindications. Priority compensation is also, however, a relatively subjective ineligibility factor. Infections and other aspects of ineligibility also probably somewhat fixed. So the most modifiable or subjective aspects are addiction and social factors and priority compensation. So eligibility has been primarily determined as predicting recurrence of harmful alcohol use. So let's examine for a moment the frequency and predictors of return to harmful alcohol use. Now there are a host of studies reporting the frequency of return to harmful drinking or return to drinking, and I've really focused on the five most contemporary and I think highest quality studies shown in reference to here. And if you take them together, you'll see there's a return to harmful drinking of just around 12%. The range is 10 to 17%. The return to any alcohol use is about 30%. So alcohol use following liver transplantation for alcohol-related liver disease is common. However, alcohol-related two to three year mortality and graft loss is very low. The overall graft loss or mortality related to post-transplant alcohol use disorder is just over 3%. All of those occurring in the Accelerate AH group. The acceptability of this is interesting because we think nothing of accepting the 14% recurrence of HCC, which is almost universally lethal. These are for patients within the Milan criteria within two years after liver transplantation. So if our focus is on utility and a survival benefit and wanting to avoid poor use of the precious donor organs that we have, we may want to think differently about the aperture within which we accept patients for liver transplantation with alcohol-related acute liver failure. Perhaps the most important potential impact of return to harmful drinking would be graft loss. And these are our data from the national analysis of the scientific registry of transplant recipients. And you'll see that alcohol-related acute liver failure has the highest proportion of graft survival among all other indications for liver transplantation. So using our current national ill-defined criteria for selecting patients for alcohol-related acute liver failure, there seems to be no discernible impact in graft survival. There have been a host of attempts to identify or predict recurrence of harmful alcohol use. I'm going to focus on two that I think are the most useful here. The first is the CPAT score, the Stanford Integrated Psychosocial Assessment for Transplant Score. This is a global assessment using 18 domains that are predictive of adherence outcomes after liver transplantation. Now, importantly, a CPAT score of greater than equal to 21 has a hazard ratio of 6.4 for relapse or return to harmful drinking, and a hazard ratio of 2.92 for rejection and 2.66 for graft loss. So it has a predictivity for very meaningful outcomes following liver transplantation. The SALT score reported by Brian Lee et al from the Accelerate AH group found that the score, which utilizes relatively narrow assessments, so greater than 10 drinks per day, alcohol-related legal issues, greater than one rehab attempt, and non-THC substance use, a score of seven or more had a hazard ratio of 2.3 for relapse or return to harmful drinking. But this had a positive predictive value of 40% for CPAT and SALT for return to drinking. Remember, return to drinking does not necessarily or rarely results in graft loss or cirrhosis. And most patients, 60% of these high-risk patients, do not return to drinking. And harmful drinking with graft loss is so unusual that predictive modeling to identify this is very difficult to derive and validate and may not be a fair assessment of eligibility in terms of utility. So I think the CPAT score of greater than or equal to 21, which will identify higher risk for rejection and graft loss is probably the most useful of the pre-transplant eligibility assessments. So in summary, how to bring it all together. What would be a reasonable approach for a patient with alcohol-related acute liver disease or liver failure? So a Madry's discriminant function of greater than or equal to 32 and non-response to medical therapy with allele score of 0.45 after seven days for the first liver decompensating event should be entry criteria. A SIPAT score which is unknown or unknowable because the patient, for example, has hepatic encephalopathy stage 3 or 4, this is a marginal candidate that really requires extensive collateral endorsement and commitment. A SIPAT score of less than 21, this is clearly a good or reasonable candidate from this perspective. A SIPAT score between 21 to 30 is a marginal but probably acceptable candidate where much of liver transplantation for this indication should be occurring. These patients will get excellent post-transplant patient and graft survival and they need intensive engagement and support with post-transplant psychiatry, social worker, regular path assessment after liver transplantation. And a SIPAT score of greater than 30, this is really at the edge of our current practice. These are, from some perspectives, poor candidates with higher risk of graft loss, rejection, non-adherence, and return to harmful drinking. And the outcomes are really unclear in this group. So on that note, I conclude and thank you very much for your attention. Hello, everybody. My name is Vijay Shah from the Mayo Clinic. These are my disclosures. Our talk today will be relating to emerging therapies for alcohol-associated hepatitis. To level set, we will start with the practice guideline of therapy in alcohol-associated hepatitis. Patients with alcoholic hepatitis should receive alcohol use disorder therapy. Individuals with mild disease should receive nutritional supplementation and treatment of complications. Individuals with severe disease defined by a MELD over 20 or a discriminant function over 32 first should be assessed for contraindications to corticosteroids. If there are none, then corticosteroids can be considered. A LIL score should be calculated at seven days and responders should continue with corticosteroids. Non-responders or individuals with contraindications to corticosteroids should be considered for liver transplantation, as was discussed earlier by Dr. Michael Charlton. Individuals who are not candidates for liver transplantation should be considered for palliative care. One of the issues with steroids is which patients may respond and which may not. This was a recent study demonstrating the maximum threshold benefit of corticosteroids residing in individuals with a MELD score greater than 20 and a MELD score less than 50, with the maximum benefit in individuals with a MELD score between 25 and 40, as shown in the dark blue region. Survival benefit is conferred up to 30 days, but not any longer than that. So clearly we need better therapies than corticosteroids. This slide shows the pathophysiology principles of alcohol-induced liver injury and sets us up to discuss some of the emerging therapies. One of the areas we'll discuss relate to liver cell death and failure and focus on regenerative therapies. Another area we'll discuss is the inflammatory response that occurs in alcohol-associated hepatitis and how to block neutrophil infiltration and immune cell activation. A third area focuses on microbe dysbiosis and increased gut permeability, and this really defines this axis of gut permeability changes, dysbiosis, leading to immune cell activation and contributing to hepatocyte death. A fourth area will focus on metabolic changes in the liver, especially parallels to NASH, which we know has a similar histology to alcohol-associated hepatitis. This pathway also leads to hepatocyte death through metabolism by cytochrome P450s. And then finally, we'll also talk about alcohol use disorder and the importance of its management in the context of alcohol-associated hepatitis. So let's start with regeneration. We know that regeneration occurs even in lower vertebrates when you cut off a digit that it grows back. And in mammals in the liver, we know that the regenerative response allows us to do liver living donor transplantation. Similarly, an acute injury in the liver will also recover. However, a chronic injury in the liver, such as occurs in the setting of liver cirrhosis, leads to abnormal regeneration and suboptimal regeneration. And in alcohol-associated hepatitis, we have not just fibrosis, but also steatosis, and sometimes the presence of corticosteroids as well, all which impair regeneration. So how do we enhance liver regeneration? There's advances occurring in bioengineered scaffolds. There's reprogramming through HNF4, a master transcription target, extracellular vesicles, cell therapies, as well as small molecule therapies and biological agents. We won't discuss all of these today, but we'll discuss some of them. This first slide focuses on F652, which is a recombinant interleukin-22, which is a regenerative therapy. A recent study demonstrated that at an increasing dose escalation and increasing disease severity that F652 demonstrated and correlated with improvement in MELD score in patients with alcohol-associated hepatitis. So this is a promising early phase study, which warrants more advanced trials. Extracellular vesicles are small particles that float in the blood and can be used as biomarkers for diagnosis, dynamic risk profiling, and importantly, these vesicles can also be loaded with therapies and delivered to individuals to facilitate treatment. And this is an ongoing area for emerging therapies as well that could stimulate liver regeneration. There are a number of bioengineering solutions that are on the horizon a little bit further out far. This includes 3D printing of livers. It includes ancillary or auxiliary livers. It also includes cells that hepatocytes, which may be obtained from pigs and other animals. So that's some of the front in terms of regeneration. What about inflammation? We know that alcohol-associated hepatitis is characterized by excessive inflammation, and this occurs through effects within hepatocytes as well as within inflammatory cells such as Kupfer cells, which can then recruit neutrophils through a process referred to as sterile necrosis. And one of the process here, the inflammasome is shown here, can be blocked through a number of compounds. And these are all undergoing evaluation in alcohol-associated hepatitis and include anakinra, canakinumab, as well as compounds which will act on the hepatocyte injury response, including seloncertib, which was a negative trial recently, as well as senacrivirac, which has shown benefit in animal models, but is yet to be tested in humans, which also can block the recruitment of neutrophils. We've done work recently showing that there are additional molecules which could be effective in blocking chemokines and their neutrophil infiltration that occurs by regulating enhancers and super enhancers, which work to produce a number of different chemokines that in turn bring neutrophils into the liver. And a number of these targets focus around the molecule transcription factor called BRD4. So that's some of the things going on in the area of inflammation for emerging therapies. What about the microbiome? We know that there's a dysbiosis and some recent trials have looked at antibiotics. This was a trial from France recently completed, demonstrating that prednisolone plus amoxicillin clovulinate had no beneficial effect compared to prednisolone plus placebo in patients with severe alcohol-associated hepatitis. We also have trials looking at fecal microbiota transplantation for alcohol use disorder, and this was a study from Virginia Commonwealth showing that individuals who had FMT for alcohol use disorder had less adverse events associated with liver disease compared to individuals who did not have FMT, suggesting a potential beneficial signal for individuals receiving FMT with alcohol use disorder and eventually with alcohol-associated hepatitis as well. So further studies will be needed in this area. Another emerging therapy is phage therapy. Phage are organisms which can kill bacteria selectively, and a cartoon is shown of the phage organism, essentially a virus-type organism, and a study recently published from the group at San Diego showing that in a humanized mouse model that phage that selectively targets certain bacteria that are most linked with the development of alcoholic hepatitis can lead to improvements in serum ALT levels as well as the histology of the liver in these animal models, and certainly this area is ripe as well for future investigation in humans. So next we'll move to the area of similarities and distinctions in alcohol-associated hepatitis and NASH. In this slide I'm going to focus particularly on altered lipid metabolism. There's a disrupted lipid homeostasis which contributes to organelle dysfunction in both these conditions, ASH and NASH. However, we don't fully understand the mechanisms by which free fatty acid lipotoxicity occurs in ASH. We do have a better understanding of this in NASH currently, and many of the drugs that we have for NASH target the metabolic pathway. This is shown in this slide looking at some of the pathways that target metabolism in NASH and whether they might have potential beneficial effects in alcohol-associated hepatitis. This is in part because there is a similar histology in both these conditions, and in fact alcohol-related hepatitis is often akin to NASH plus increased cholestasis, and indeed compounds such as FXR agonists are being tested in NASH and some initial studies going on in ASH as well, but they will need a more rigorous analysis in bigger studies. But you can see that there's a number of pathways that might be relevant even outside of the FXR pathway that could potentially be beneficial for alcohol-associated hepatitis as well. This may include PPAR agonists, FGF21, FGF19, and a number of other molecule targets that are now being rigorously evaluated in NASH as well. So now during the last portion of the talk, let's pivot to strategies to mitigate the impact of alcohol-associated liver disease at global, local, regional, and individual levels. And the next several slides will focus on the interrelationship of alcohol use disorder and alcohol-associated liver disease. So that we know that at a global level, we need policy changes that are society-wide that can impact alcohol consumption. This includes pricing strategies, efforts to reduce alcohol availability, and restricting alcohol use to groups so as to reduce the amount of alcohol consumed. And this may include taxation, minimum unit pricing, regulation of hours of sale, marketing approaches. Cascading down from the global level is the regional level. Here it's important to identify alcohol use disorder and alcohol-associated liver disease and the coexistence of both. This may include population screening, outreach programs, and multidisciplinary approaches, which we'll talk more about. We know there's a number of screening approaches that we can use to screen for both conditions. There are outreach models using digital technologies and telehealth. And an example of this is Project ECHO that was used for hepatitis C. Finally, at the individual level, we need individualized treatments for alcohol use disorder and alcohol-associated liver disease. And here, again, a number of therapies that we know about for alcohol-associated liver disease, which we went over for hepatitis. Alcohol use disorder therapies, including pharmacotherapy, motivational enhancement therapy, cognitive behavioral therapy, and brief interventions. And then complementary models, which we'll talk about. So this is an example of an integrated therapy for alcohol-associated liver disease and alcohol use disorder, putting the patient in the middle, surrounded by the gastroenterologist, hepatologist, the primary care physician, and then the patient's home using chronic care models, technology, and approaches for multidisciplinary care. And indeed, a recent study from our group showed that individuals with alcohol-associated hepatitis who are discharged from the hospital have improved survival and less rehospitalization if they attend alcohol rehabilitation prior to leaving the hospital or very soon after. So this seems like a good model. And so what are some of the key recommendations? And most importantly, what are the barriers that we need to overcome? Well, the key recommendations include a co-located team in multidisciplinary management, strong interpersonal team relationships, use of novel patient encounters, and outreach and educational activities. Potential barriers include the financial sustainability of this model, logistical complexity, disparities, and the cognitive status, especially patients who may have hepatic encephalopathy. So what's on the horizon? Well, we know digital biomarkers of alcohol use disorder might be quite important to predict craving and relapse. And this could be done through cell phone models. We have studies ongoing now looking at dynamic behavioral data that we can obtain through a cell phone, active and passive data through registered individuals who are looking to help us help them with their alcohol use disorder. So as I start to wrap up here, I'll talk about the unmet needs in the field. These include the management of alcohol use disorder and alcohol-associated liver disease, including the ideal teams and study endpoints. They include digitization, automation, and smart processes for diagnosis, prognosis, and new treatment pathway algorithms, concordance of the clinical syndrome of alcohol-associated hepatitis with the histologic lesion of steatohepatitis, and differentiation between infection versus sterile inflammation and local hepatic versus systemic inflammation. And what's down the road? We know that we will develop artificial intelligence algorithms for histology, radiology, and individualized therapy. There are therapies I mentioned relating to the microbiome and phage therapy. I also mentioned biological therapies and gene therapy, perhaps immunotherapies that can target T cells like we do for other cancer conditions currently, and then small molecule inhibitors, as I mentioned, that could target super enhancers and epigenetic targets. So these are the key takeaways of my talk as I wrap up here, that the benefits of corticosteroid therapy are limited to select groups with a MELD score in the range of 25 to 35 is the best range for therapy and limited duration of 30 days. Efforts to stimulate liver regeneration and reprogramming are encouraging. Studies are ongoing to target the inflammasome and inflammation. We have a need to integrate alcohol use disorder therapy with alcohol-associated liver disease therapy, and some therapies for NASH should also be evaluated in alcohol as well. Thank you very much. Good afternoon, and thank you to the conference organizers for the opportunity to talk to you today about bariatric surgery in youth. In the next 20 minutes, we'll be covering indications, outcomes, and ethical considerations. Here are my affiliations. Most relevant to this talk, I have served as the Medical Director of our Surgical Weight Loss Program for Teens for the past 15 years, and here are my brief disclosures. Over the past three decades, bariatric surgery has become an accepted evidence-based treatment for adolescents with severe obesity. Most of the increase in the rates of surgery in youth occurred around 20 years ago when there was a four-fold increase around the year 2000. Since that time, rates in adolescents have plateaued, and we'll talk a little bit more about potential factors behind that later on in this talk. So first, how do we even get here where we're offering bariatric surgery to children? Well, severe obesity is a major problem in the United States. Currently, about 6% of youth have class 2 pediatric severe obesity, shown here in red. Two and a half percent have class 3 pediatric severe obesity, and when we drill down to adolescents, it rises to 10% with class 2 obesity and 5% with class 3 obesity. So how do we define classes of obesity in youth? What you see here is an extended BMI chart. So unlike adults, where we can use a static BMI cutoff, because of normal growth in children, we have to use percentiles. So overweight is classified as a BMI, which is between the 85th and 95th percentile for age, and it's sex-specific for boys and girls. Obesity is classified as a BMI over the 95th, equal to or over the 95th percentile, shown here in this range. To classify severe obesity, we use percentiles of the 95th percentile. So 120%, or I like to think of it as 1.2 times the 95th percentile BMI, is considered class 2 obesity, and for class 3 obesity, it's 140% of the 95th percentile. And what you'll know when you look at ages is that often this BMI is below the BMI of 35 or 40, which is what we classically use to consider severe obesity. Here's an example. So if we look at a child with a BMI of 30, that may not raise any red flags in your clinic unless you pull up one of these extended BMI charts. When you look at a child who's age 11 and a half, a BMI of 30 is actually at the 120% of the 95th percentile line. So this is a child with severe class two obesity, although the BMI itself may not raise a lot of eyebrows because it seems like it's consistent with obesity. So severe obesity is also a problem worldwide. We can see here in red, which designates severe obesity, that this has been gradually increasing over the last 40 years worldwide, including in regions of the world where severe obesity was never a major issue. In fact, undernutrition or malnutrition, shown in blue and purple, were the major issues. So you see this in Middle East and North Africa, Latin America and East Asia, as well as Southwestern Europe. And unfortunately, children with severe obesity are at much higher risk of severe metabolic problems. This chart shows the increase in odds of having type two diabetes in severely obese children compared to those with overweight alone. And you can see for both females and males with class two obesity, the odds are 15 to 19 fold higher. And for those with class three obesity, it's 25 and 39 fold higher than an overweight adolescent. And the same has been shown for increased risk of prediabetes compared to class one obesity, that there's an increased odds of prediabetes. And many centers across the United States and worldwide now have clinics focused on youth with type two diabetes. So what about fatty liver? We know that the prevalence has also increased as the severity of obesity worsens. So if you look at all U.S. children combined, about 10% have fatty liver. When you look at children with overweight or obesity, the prevalence rises to 30%. And when you look at adolescents presenting for bariatric surgery, we're seeing rates as high as 60%. Also when we look at cohorts of children enrolled at centers, NASH centers across the United States, the mean BMI is 33 to 34. And as you recall from the extended BMI chart, that's a BMI range that's clearly within the severe obesity range for the vast majority of children. We also believe that severe obesity may be associated with a higher risk of more histologically severe or advanced NAFLD. We did a single center retrospective analysis in Cincinnati, and we found that increasing classes of obesity were associated with increased ALT and GGT levels. When we looked at the subset that had had either MR elastography or biopsy confirmation, we likewise saw that there was increased liver stiffness and a greater proportion with NASH scores above five in the class three obesity group. Unfortunately, children with severe obesity are very resistant to conventional and pharmacological treatments for obesity. Mean weight loss tends to be under 5%, both for dietary and physical activity interventions, as well as for some of the medications that we use to treat obesity. Sibutramine had the highest mean weight loss, but was pulled from the market due to post-marketing adverse effects. Liraglutide was just recently studied and published, and you can see that midway through the one-year study, the mean weight loss approached 5%, but then drifted back to 3% mean weight loss, potentially because of the daily injections required, which can be difficult for some children to adhere to long term. But so far, this is one of the most promising medication treatments we have for severe obesity, and even that tends to result in minimal mean weight loss. So therefore, for over a decade, there have been established criteria for considering a bariatric surgery in youth. Here are the most recent published guidelines, where above here for class 1 obesity, it's considered appropriate to consider this if you have a child with a severe comorbidity defined as moderate to severe sleep apnea, type 2 diabetes, pseudotumor, or severe and progressive NASH. For those with class 3 obesity, any obesity-related comorbidity or self-reported quality of life impairment. The most recently published NAFLD guidelines from ASLD in 2018 and from NASMHIN, which is our pediatric GI society, they felt that it was premature to consider bariatric surgery in established treatment for NASH because there was a paucity of data. Now, that's changed, and so these guidelines may be updated in the near future, but for now, it was considered appropriate for youth with NAFLD that otherwise met the above adolescent bariatric surgery criteria. In addition to these medical criteria, for youth, it's important to consider additional psychosocial factors. Behavioral factors include readiness to lose weight, results of prior attempts, adherence to follow-up, family and social support, and whether they have any active psychiatric disease or substance abuse concerns. Likewise, it's important to consider the decisional capacity of adolescents. Oftentimes, adolescents are brought to the programs by their parents, so it's important to understand if the adolescent is facing any coercion. Sometimes, the youth doesn't want surgery, even though the family does. Are they willing to comply with the needed diet and behavioral modifications, and do they have sufficient emotional and psychological maturity? In addition, many pediatric bariatric surgery programs see children with very unique coexisting medical conditions, children that have limited mobility, children that have syndromic obesity or hypothalamic obesity, children with developmental delay. Also, we sometimes get referrals for much younger children. In some countries, even as young as five have had surgery, and these are children typically that have severe sleep apnea or severe skeletal abnormalities that are significantly impairing their quality of life. For all of these groups, there have been some preliminary promising data, but we're going to need more rigorous, prospective study of outcomes. For these children, it's important to do case-by-case evaluations with careful consideration of risk-benefit ratio to determine if the timing of surgery is appropriate at a younger age. It's also helpful to consider an ethics consult, and often, a further evaluation and more extended preparation may be required. Typically, it takes about three to six months to get a youth ready for surgery, but in these unique situations, it may take longer, a year or more, to observe and make sure that the child is ready. What are the outcomes in youth? Do they differ from adults? As in adults, the primary operations used in adolescents include runewide gastric bypass and vertical sleeve gastrectomy. Over the last 10 years, vertical sleeve gastrectomy has become the predominant surgery performed in both adults and adolescents. Our best long-term data in the United States comes from the PROSPECTIVE NIDDK-funded multi-center teen lab study, which enrolled adolescents at six U.S. centers with rigorous measures and independent adjudication of complications. At baseline, the study enrolled 242 adolescents with mean age of 17, median BMI of 51, and the majority were white and female. As you can see, during the course of enrollment, the type of procedure shifted to sleeve gastrectomy in the medium color gray, whereas the use of bypass declined. Banding was done only in a small minority of patients because this procedure was never approved by the FDA for use in adolescents in the United States. At three years, teen labs reported significant and sustained weight loss after both sleeve gastrectomy and bypass. When looking at patients who achieved a 10% or higher BMI reduction, the rates were high after both bypass and sleeve. There was perhaps a trend for slightly higher proportion in the bypass group, 89%, but because of the small numbers of sleeve patients, we couldn't directly compare these two groups statistically. Likewise, only 2% of patients after bypass returned to their baseline weight, so had significant weight regain, whereas 4% after sleeve had significant weight regain at three years. In comparison, when we look at the Liraglutai clinical trial just published in 2020, only 24% of adolescents achieved a 10% or greater weight loss. That just shows you again how dramatic the weight loss is after surgery. Among teen labs participants, there were very high rates of comorbidity resolutions in particular type 2 diabetes, but also other significant comorbidities like dyslipidemia, elevated blood pressure, and abnormal kidney function. In terms of fatty liver, the study wasn't designed to rigorously assess fatty liver, but 157 patients did have intraoperative liver biopsies, with 60% having fatty liver and 20% having either borderline or definite NASH. In general, fibrosis was mild, with most of the patients having none to stage 1, whereas in most of our NASH clinics, we're seeing about 30% to 40% of patients with stage 2 fibrosis and 10% to 15% with stage 3, so this may not be as representative of the patients that we see in our NASH programs. Nonetheless, across all groups, there was a decline in ALT. Shown here is the definite NASH group in dark blue compared to NAFLD and borderline NASH in yellow and green, and patients without fatty liver, and you can see that the definite NASH group, although they had a significant decline in ALT, the mean ALT approached 40 and then trended up a little bit at 3 years, so I think this highlights that we need longer term studies to determine if patients with more severe disease are going to have sustained remission and improvement of their NASH over time. Likewise, we have very limited data on histologic outcomes. This was a single-center study from Italy of 20 patients. Only six of these had definite NASH. All of the patients had significant improvement in NASH score, and the majority had mild fibrosis, of which most improved, compared to a lifestyle group, which essentially remained stable with no improvement in NASH or NAFLD. Now, again, this is a small single-center study. Only a minority had NASH. The type prevalence of type 2 diabetes wasn't known. We've just completed a similar study at our site and are in the process of analyzing those results and plan to report those within the next year, and our cohort has a higher proportion with NASH and diabetes, so it will be interesting to compare those results. Does age of surgery matter? Should we be waiting until children or older adolescents? We looked at this in teen labs by dividing the cohort into young adolescents, 13 to 15, versus older adolescents, and we found comparable BMI change shown here on the left graph. In terms of diabetes remission, there was a trend for older teens to have higher remission rates, but the younger teens also had more SLEE procedures, and that may be something significant that we'll talk about in a minute. There's been another rigorous prospective study of bypass in Sweden, and they compared adolescent timing of surgery to adults who received surgery, and they found that both adolescents and adults had comparable rates of BMI decrease, shown here in red and green, compared to adolescents enrolled in a lifestyle intervention. When broken down by the proportion attaining significant weight loss, you can see that the adult patients in green were shifted more towards higher proportion achieving significant weight loss, whereas the red adolescent group was a little bit more flattened with some of the participants attaining less than 20% weight loss. Does type of surgery matter? In adults, emerging data has shown that bypass is associated with greater weight loss and remission of type 2 diabetes at five years out from surgery. We don't yet have similar published studies in adolescents, but this is an area of important interest and study, but again, bypass has higher operative nutritional risk, so this needs to be factored in. This graph shows that adults off of diabetes medications was higher after bypass versus SLEE. Comparable operative complications have been reported in adolescents, so it doesn't seem to be higher risk to do the surgery in youth. In teen labs, there were no deaths and major complications were 8%, minor were 15% in the first 30 days. Long-term, SLEE had a lower 13% prevalence of reoperations and procedures compared to bypass at 25%. Most of these reoperations were cholecystectomies or endoscopies for reflux or stricture dilation. And SLEE also has a lower risk of significant nutritional deficiencies, with only 20% after SLEE having two or more deficiencies compared to 60% after bypass having two or more deficiencies. The vast majority of deficiencies, the primary deficiency was iron, again, twice as high after bypass versus SLEE, less often B12 and vitamin A deficiency, which was only an issue after bypass. Vitamin D was high pre and post. Again, adolescents are less adherent to supplements over time, so importance is shown by opening of their nutritional supplement caps, declining rapidly in the first six months after surgery. So it's important to follow these young people as they age into adulthood, and particularly the women as they may become pregnant. Again, suboptimal weight loss does occur as about 10% to 20% of teens have less than 9% reduction. When looking at teen labs and dividing it into those that lost less than 20%, shown here in the upper line, and those that lost more than 20% over five years, shown with the dotted line, we found similar metabolic outcomes in general, except for dyslipidemia and high blood pressure, which are more likely to be improved if you lost more weight. So what this shows is that the metabolic benefits for the majority are maybe weight loss independent. Postoperative loss of control eating was the only post-op risk factor associated with suboptimal weight loss, and we really don't have any preoperative factors that identify those at greater risk. But lastly, it's important to remember that only a minority of eligible adults and adolescents undergo bariatric surgery. Just over 1% of medically eligible adults underwent bariatric surgery in 2018, and the overwhelming majority of these procedures were done in adults with less than 1% performed in adolescents. So why is this? Well, there's a multitude of factors. The procedure is expensive, insurance coverage remains variable, there are more limited programs willing to do surgery or comfortable doing surgery in adolescents or youth, there may be language barriers for some of our minority patients, there may be under-referral of eligible patients, and I will say as somebody who's practiced in this area for 20 years, there are definitely patients and families who, despite a good medical eligibility and probably high likelihood of success, are just not interested in having a surgery done. Likewise, we've seen racial and ethnic disparities. Above here on the top bar are shown the expected distribution of race and ethnicity as one might expect if you looked at just severe obesity rates in youth populations, and down below is what we actually observed. We saw more white and fewer black or Hispanic patients receiving surgery. We don't completely understand why this is the case, this is an area of active research, but at least one preliminary study suggests it wasn't related to public insurance status, so there may be complex factors underlying these disparities. So in summary, the three take-home points I'd like you to leave with are that, one, bariatric surgery is the most effective means that we currently have to achieve durable weight loss in adolescents with severe obesity. Data obtained over three to five years suggests very high resolution rates of both type 2 and NASH in youth after bariatric surgery, with perhaps a trend towards higher resolution rates after bypass, but more data are needed. Unfortunately, only a minority of eligible youth are referred and receive surgery. So in my view, the main research gaps we need to focus on are further understanding long-term outcomes 10 years or more after surgery, to understand are there optimal surgery types for specific conditions, and how long do responses last. In addition, I think we need to do more work to understand barriers to access and see if any of these might be modifiable. And with the emergence of endoscopic bariatric treatments and improved weight loss medications, including the GLP-1 receptor agonists, it may be that we integrate some of these treatments to extend the range of options available to youth with severe obesity and also to help youth who have suboptimal weight loss outcomes after bariatric surgery. So with that, I'll end here, and I'd be happy to take your questions during the open-question-answer period. Thank you. Hello, I'm Dr. Steven Harrison, and I would like to thank the organizers of this postgraduate course for the incredible opportunity to present to you today on emerging therapies for NASH, a very, very hot topic. Here are my disclosures. The objectives of today are fourfold. We'll review the current endpoint assessment and rationale. We'll review the drugs in development in phase 2 and phase 3. We'll look to combination therapy, and then end with some future considerations that I hope are provocative and engender thought. So I wanted to start with where we were 10 years ago and kind of juxtapose against that of 2011 to now. So prior to 2010, there were about 64 NASH studies in phase 2 or 3 development and no cirrhotic trials. Back then, the earliest study we had published of any significance relative to drug development was the PIVENS trial published in the New England Journal of Medicine in 2010. Back then, histopathologically, we were looking for improvement by one or more points in hepatocellular ballooning score, no increase in fibrosis, and either a decrease in the activity score for non-alcoholic fatty liver disease to a score of three or less or a decrease in the activity score of at least two points. Fast forward to studies since 2011, we see 254 trials in phase 2 or 3 with a whopping 38 NASH cirrhotic trials, and this comes from a clintrials.gov search done in October. More recently, the FLINT trial published in Lancet by Nushwander Tetri and colleagues histopathologically looked for improvement in centrally scored liver histology defined as a decrease in the NAFLD activity score by at least two points without worsening of fibrosis from baseline to end of treatment. And then more recently, we can look to the liver form in a publication in Hepatology in 2019 that defined that a little bit better, looking at resolution of NASH without worsening of fibrosis and or improvement in fibrosis without worsening of NASH as an endpoint, and resolution of NASH here was defined as a disappearance of ballooning and a disappearance or resolution of inflammation. So where are we at today? Today in 2021, the FDA requires a histopathologic endpoint that we establish as subpart H approval or a surrogate for long-term outcome benefit. And those two potential options are NASH resolution and or fibrosis improvement, and you see the criteria associated with them. So how did we get there? Well, ultimately, NASH progresses to long-term outcomes that are negative in a certain number of patients, but this could take years, even decades. So to develop drugs in NASH, the FDA looked at a potential surrogate outcome. The histologic endpoints of NASH resolution and fibrosis improvement. What's the evidence for this? Well, NAFL liver fibrosis is a risk for adverse outcomes, and this is just a representative slide of several different studies that have looked at this, both with severe liver disease and liver-related mortality. And you can see on the left, really beginning to ramp up at F3 disease, and on the right, liver-related mortality. We see that begin to ramp up at F2 or greater disease. Now what about regression of fibrosis? Is regression of fibrosis associated with improved clinical outcomes? This study presented by Arun Sanyal is now in hepatology and press suggests that is exactly what's happening. On the vertical axis, you see liver-related events, and here on the horizontal axis, you see NASH-CRN scoring or ISHAC scoring. Either way, if you look at the green bar, improvement in fibrosis stage, you see a lower liver-related event percentage compared to no improvement in fibrosis. This is very, very helpful. Now, what about fibrosis regression over time? Well, here's an example of bariatric surgery in patients with severe obesity and NASH. On the left-hand pie chart, you see lots of green and lots of yellow. That's NASH with mild to moderate inflammation. As you progress from one year to five years post-surgery, you see more and more of the no NASH part of the pie. It turns out that NASH resolution with no worsening of fibrosis was seen in 84% of patients five years post-surgery, and fibrosis improvement was seen in 70% of patients five years post-surgery. Furthermore, when we look at is NASH reversible and hone in on that question, we see this additional analysis of this bariatric surgery study with 180 NASH patients. At five years post-surgery, as previously mentioned, 84% had NASH resolution with no worsening of fibrosis. If you look at the bars on the left-hand bar graph, resolution of NASH according to weight loss, you can see that when you focus on BMI loss of zero to five, five to 10 or greater than 10, you see gradations of increasing amounts of NASH resolution without worsening of fibrosis. In the red bars to the right, you see that as you go from baseline to five years, the percentage of none to mild fibrosis increases significantly, noting this association between NASH resolution and fibrosis improvement. Furthermore, Dave Kleiner and colleagues recently published in JAMA, a similar type result that showed on the graph here on the left, vertical axis fibrosis stage change and on the horizontal axis change in the NAFLD activity score. As you go from right to left and you improve the NAFLD activity score, you actually decrease fibrosis staging. On the right, we see fibrosis stages and patients improving from having NASH on the first biopsy to having no NASH on the second biopsy and you see the percentage of none to mild fibrosis increase. This data has been further supported by presented but yet unpublished data from the GOLDEN-505 trial, which is a trial with the drug Elifibranor. Here you see on the left, as you improve ballooning and inflammation from right to left, you see fibrosis improvers as noted by the orange bars. Conversely, as you worsen inflammation and ballooning, in green bars, you see fibrosis worsening and this is also exemplified on the graph on the right. So we need to focus for sure, but let's not lose sight of the full picture. We are taking care of a person, not just a liver. So while the potential targets for drug development are focused on NASH resolution and fibrosis improvement, we can't forget about the lipotoxic fat and the need for patients to lose weight often and atherogenic lipid improvement and glycemic control. And ultimately what we're aiming for are improvements in major adverse liver outcomes, improvement in major adverse cardiac events and reduction in overall mortality. So the question you would ask is, is a single therapy enough? Well, we know now, dating all the way back to the original two-hit hypothesis by Chris Day and colleagues, that this is a complex multifactorial disease. This opens the door for multiple potential therapeutic targets, many of which you see here. Early on in development, we focused on monotherapy, but as you'll see from this presentation, we firmly believe that ultimately treatment will be multifactorial, multimodal and combination in nature. So here are just five buckets I like to put drug development in. Targets related to insulin resistance and or lipid metabolism, targets related to lipotoxicity and oxidative stress, targets related to inflammation and immune activation, those related to cell death and then targets focusing on fibrosis. As you can see, the majority of drugs developed to date are targeting those first two buckets, insulin resistance, lipotoxic oxidative stress. In black are drugs currently in phase two, in red are drugs currently in phase three. Now, over the next several slides, I'm going to jump into this a little bit more detail. First off, let's look at those drugs in phase two. Here, I broke them into non-invasive endpoints and histological endpoints. Another way to look at this is drugs on the left are in earlier phase development, like phase 2A. Those on the right are more paired liver biopsy phase 2B studies. So let's first look at the non-invasive endpoints on the left. Here, we see seven different categories, multiple drugs in development that are FXR agonist. We see multiple drugs in development that are using a GLP-1 receptor agonist as base and then adding either a glucagon agonist or a glucagon agonist or a GIP on top of that. You can see that there's lots of different drugs using different mechanisms of action being developed early on. As we shift to the right-hand table, here we have phase 2B drugs looking at histopathologic endpoints. If I have a green box around it, those studies are ongoing and continue to enroll. If there's an arrow, a green arrow, then those studies have already read out or finished enrollment and are waiting for results to be read out. I noted three asterisks by NGM and Alda Fuhrman and BMS's Peg Belferman because both of those will be described in detail at this liver meeting. Now looking at agents in phase 2 development, in particular, separating out the route of administration. On the left, we have all the oral agents and their mechanisms of action and the company that's sponsoring them noted on the left. On the right, we have those that are injectable or infusion therapy, and you see those are FGF19 agonist, FGF21 agonist, GLP receptor agonist, DGAT2 inhibitor, and galactin 3 inhibitor. Where are we at with phase 3 development? We've seen an explosion of drugs making their way to phase 3. We currently have five drugs in phase 3. One is currently on hold, but abetacolic acid has already completed its enrollment of subpart H. As you guys know, there was a complete response letter sent after filing the new drug application, but Intercept is aggressively trying to reevaluate additional data for both efficacy and safety and hope to refile in the near future. Lanofibranol recently started enrollment in its phase 3. Resmeterone has two phase 3 trials ongoing. Maestro-Nafl-1, which is a non-invasive study, looking at safety and efficacy of non-invasive tests, and then Maestro-Nash, which is the registration trial, and those trials are still ongoing. Aramcol is currently on hold, and semaglutide recently began its phase 3 trial in April of this year. So looking next at a couple different slides focused on the Nash landscape and these histopathologic endpoints, we'll start with fibrosis improvement first, and I focus on compounds in phase 2 development. Those are all in the brown boxes and highlighted by the orange bars below. We see placebo response rates for fibrosis of 18 to 20%, overall drug response rate of 24 to 62%, depending on mechanism. The highlight here is that not much is dependent on time. I disagree with my colleagues sometimes here. I don't think time is always necessary for fibrosis improvement. It more depends on the mechanism, but that's still controversial. When looking at drugs for fibrosis that are now moved on into phase 3, these are not the phase 3 results. They're the phase 2 results that propelled them into phase 3, and here we see placebo responses of 12 to 33% and drug responses of 18 to 48%. When looking at Nash resolution, let's go back to those compounds in phase 2, and take a look at what we're seeing here, and here we see placebo responses of 8 to 9% and 19 to 54% drug response. When we're looking at phase 3, we see a placebo response of 5 to 22% and then a 7.5 to 59% overall drug response. Now, shifting to both Nash resolution and fibrosis improvement, this is a much tougher endpoint to reach, but you see response rates here of compounds in phase 2 of up to 37%. In phase 3, or drugs that are now moving into phase 3, 35 to 37%. So now let's focus on Nash cirrhotic trials. There are currently five studies that are underway. The beta-colic acid trial has completed enrollment and is just awaiting the results, and you see the other drugs currently enrolling, even a combination therapy trial. Now, what about the future of Nash therapeutics? These are where we are with the ongoing phase 3 trials, and we see that a ram call is currently on hold. Resmeterone part one surrogate endpoint is expected in the Q3 of 2022 with long-term outcome in March of 2024. A beta-colic acid, again, looking at that long-term outcome data in 2025 on top of a potential refiling. Semaglutide part one, April 2024, that's the surrogate endpoint readout, and you see the outcome out in 2028. Lanofibranol, very similar, about February 2024 for a surrogate endpoint, and then outcomes in 2028. So many drugs have been tried and many have made it to the graveyard. We see those listed here, but there have been incredible lessons learned from these trials that we've been able to apply prospectively to new studies, and each time something like this happens, we learn from it and get better the next time around. Now, this is my perspective. It's only my perspective and my opinion on where we are. This is a little bit complicated, but I wanted to show in this graph the histopathologic effect, the extrahepatic benefit that we're seeing relative to its adverse event profile. Adverse event profiles that are mild are in orange, mild to moderate are in green, and you can see as you move from left to right histopathologically where drugs stand relative to their benefit, and then from bottom to top how well they're affecting extrahepatic benefits, whether that's atherogenic lipids, glycemic control, or weight loss. Ideally, the best drugs will be in the upper right-hand corner of each. I broke these into non-cirrhotic NASH and cirrhotic NASH. For example, recent drugs that have moved on to phase three, semaglutide, you see that it's about halfway on histopathologic effect, does really good on NASH resolution, not so much on fibrosis improvement, although there is data that it stabilizes fibrosis progression. But it does have effects on many of the extrahepatic benefits that we're looking for in a drug, such as weight loss, liver fat content reduction, atherogenic lipids, and glycemic control. Moving on to where we're going, I like to look at this through the lens of an airplane. Right now, we're at the biplane stage, but through scientific development and further refining of our understanding of NASH pathogenesis, we will eventually get to that fifth-generation jet fighter you see on the right. We're just not there yet. So looking at combination therapy, which again is, I think that's the wave of the future, I break them into two different parts, those with milder disease and those with more advanced disease. I think that patients that have less advanced fibrosis, we need to really get after agents that target their metabolic profiles, limiting liver fat, transitioning into the liver, and shutting down inflammation and ultimately cell activation. Alternatively, if you've already become advanced, we need to halt disease progression. The lines I have above that balance there is really if we're targeting milder disease, safety is paramount and tolerability is paramount because these patients don't have a lot of symptoms and we need to carefully help their livers remove fat, down-regulate inflammation. On the other hand, if you're advancing towards cirrhosis, we need to get after that disease in a major way and potentially our risk tolerance level is a little higher. So the ideal combination therapy is one that's oral, well-tolerated and safe, synergistic in that it improves histopathology and extra hepatic metabolic profiles and enhances long-term outcomes. Here is just a sampling of many of the trials that are currently underway today looking at combination therapy. So what about the future? And this is where I want to spend the last remnant of my time with you today. I think we are going to get to the point where we can have this addition of precision medicine to attack fibrosis, for instance. And on the right, I illustrate a very elegant body of work that was done by Amler and colleagues and published in Nature in 2020. In fact, Scott Lowe's group was also involved and Scott Friedman participated in this work as well. And it has to do with these senolytic CAR T-cells reversing senescence-associated fibrosis. And here in this example, we see a mouse model, a carbon tetrachloride-induced fibrotic mouse model that was treated with these CAR T-cells, these senolytic CAR T-cells. And after 20 days, the livers were harvested and you can see significant down-regulation of collagen as well as ALT and even AST. So there is potential future there. Now, what about these therapeutic oligonucleotides? Here's some work that was recently, a review that was recently published by Naeem Al-Khoury and his group looking at liver metabolic targets of these therapeutic oligonucleotides. And some of these are in development today. The siRNA against heat shock protein 47, the antisense for DGAT2 and antisense for PNPLA3, just to name a few. And then of course, genetic polymorphisms are also important, PNPLA3 and HSD17 beta-13 and targeting those potentially could also be helpful. For histological endpoint challenges, we have a problem currently with our histologic reading, sampling variability, inter- and intra-observer variability, various central reading processes across programs, different scoring systems, core size, length and staining variability, and oh yes, even placebo response as highlighted here on the right, where you have a beta-cholic acid, 25 milligram and L-ethybranol. Both of those drugs had similar fibrosis improvement in their phase three trials, but the placebo response killed L-ethybranol while as a beta-cholic acid is still trying to attain approval. So this has to do with the suboptimal reliability of liver biopsy evaluation as I published with some colleagues this past year, showing that the overall inter-reader comparisons is quite low. In fact, agreement between three pathologists is less than a flip of a coin. It's around 12% for the NAFLD activity score, while it's a modest 69% for NASH diagnosis. So how can we improve our endpoint assessment? Well, I think histopathologic requirements, if they're still required moving forward, we need to look at more tissue. Why do we only look at one H&E and one trichrome? Why not three H&E stains, for instance? It's three times as much liver tissue, and the liver is telling us we need to be looking at a broader piece of tissue. We need to refine potentially our NASH diagnosis. Do we need to modify our ordinal fibrosis scoring system? What about bringing in things like ductular reaction or portal inflammation, which has been linked to fibrosis? More than one pathologist, I suggest utilizing a panel review, two readers plus one adjudicator. But there are different ways to look at this. AI digital pathology to augment ordinal scale pathology reads could also be helpful, and consider using the SAF instead of the NASH CRN histopathologic interpretation. On the right, I show you some provocative new data using AI digital pathology that now has shown that a 17% improvement in collagen is linked to a one-stage improvement in fibrosis by an ordinal scale. Furthermore, here, this idea of using the SAF instead of the NASH CRN is not new. Here's work published previously showing that the percent of biopsy interpretation concordance with reference evaluation increased from 77% to 97% and from 42% to 75% after the use of the SAF score, and that here, the SAF score should decrease inter-observer variation among pathologists. Now, we can also focus on development of NITs linked to long-term outcomes. I list three here, ELF, MRI, CT1, and MRE, and I show you just some examples of where ELF score cutoffs could potentially be predictive. Here, we have CT1 showing that 840 milliseconds is predictive of event-free survival and all-cause mortality. And then finally, MR elastography, where we see MRE now able to predict long-term outcomes, as illustrated here. Now, turning back to the SAF versus the NAFLD activity score, here you see with the NAS, all features are combined and it's not diagnostic, whereas with the SAF, it's separate assessment and there is a diagnostic score, and you see the breakdown for ballooning and lobular inflammation, but these are not interchangeable scores. A recent study, actually a study done back in 2017, looked at NAS and SAF evaluation in a set of 1,000 liver biopsies. And here, while there was excellent concordance for definite NAS between both scores, when you had a borderline NAS case with an NAS of three to four, there was only 88% of the time, or 88% of the time, you had borderline NAS diagnosed as NAS, so suggesting that the SAF might be a bit more consistent over time. And then finally, we have an example with lanofribinol that I showed previously, but now this trial has recently been accepted for publication in the New England Journal of Medicine. And what we see here, by intent to treat, is a reduction of at least two points of the SAF activity score, no worsening of fibrosis. So in this large Phase IIb trial, they did not use the NASH-CRN, but rather use the SAF, and you see a nice dose-response relationship. I'm not saying this couldn't have happened with the NAFLD activity score. What I'm saying is that it's okay to use the SAF in your Phase IIb, and then transition back to the NASH-CRN for your subpart endpoint with the FDA. So in summary, five bullet points. As our understanding of NASH pathogenesis improves, the number of targets in NASH drug development continues to expand. Therapies aimed at both histopathologic and dysregulated metabolism are ideal. Synergistic combination approaches, potentially overlaid with precision medicine advances, may allow for broad application. Current regulatory endpoint challenges exist, but I believe they're not insurmountable with potential modifications. And data supporting NITs that demonstrate long-term outcome benefit are growing and offer hope for usurping liver biopsy-based endpoints. Thank you for your attention.

liver transplantation

alcoholic hepatitis

early identification

referral

corticosteroids

aggressive listing

transplant eligibility

NASH

non-alcoholic steatohepatitis

drug development

histopathological endpoints

insulin resistance

fibrosis improvement

combination therapy

precision medicine