Welcome to the AASLD webinar, NASH as the cause of hepatocellular carcinoma. I am Fasiha Khanwal. I'm a professor of medicine at Baylor College of Medicine in Houston, Texas. And I will be moderating the webinar today. The objectives of the webinar are to learn about epidemiology of NAFLD or NASH-related HCC, differences in NASH HCC compared to HCC from other etiologies, and to talk about prevention and management of NASH HCC. As you can see, we have a great lineup of excellent speakers. They will discuss the rising burden of NASH HCC. They will also review clinical manifestations, prognosis of HCC, and describe data on screening, prevention, and treatments, and give you some of the novel biomarkers and novel treatments that are in the pipeline. Before I introduce our first speaker, a quick reminder, we heard about it. The speakers will take all the questions from you at the end of the webinar. So please provide any questions that you have via the Q&A chat box function, and we will go over them at the end. With that, it is my pleasure to introduce our first speaker, Dr. Zubair Yunusie. Dr. Zubair Yunusie is the president of Inova Medicine as well as professor and chairman of the Department of Medicine at Inova Fairfax Medical Campus in Virginia. He specializes in hepatology and gastroenterology and has authored over 560 articles with some of the leading research in epidemiology of chronic liver diseases, including HCC. He will be reviewing some of these data with us today. Thank you, Zubair, for being with us today, and over to you. Thank you very much, Dr. Kahnewald, and thanks to ASLD for allowing me to participate in this great meeting. My task is really to give you about 15 minutes of overview on the epidemiology of non-hepatitis in the context of HCC. These are my disclosures. Let me just remind you that the global prevalence of non-chronic liver disease is about 25%, with the highest prevalence being reported from the Middle East and South America. When you look at patients with type 2 diabetes, the prevalence is about 55-56% globally, and diabetics, again, very high prevalence in the Middle East, as well as actually Europe and the rest of the world. Of course, an important group that we would always worry about would be children. With NAFLD, the prevalence is about 8% globally. When you look at the spectrum of NAFLD, you're talking about those patients that have just fatty liver without any evidence of liver cell injury to those that have steatohepatitis or non-alcoholic steatohepatitis or NASH. This is a diagnosis that requires a liver biopsy. Of course, that's what steatohepatitis is, and it is the progressive form of NAFLD. When you look at the prevalence of NASH in the general population, you just can only estimate the prevalence rates in general population, somewhere between 1.5% to 6.5%. Again, when you look at type 2 diabetics where you have a very high prevalence of NAFLD, you also have over a third of these patients having non-alcoholic steatohepatitis and at risk for progressive liver disease. Now, the second part of the burden of NASH and NAFLD, of course, is progressiveness. It's important to remember that there are a number of risk factors that contribute to development of NASH, and they are also going to be important as risk factors for HCC, obesity, type 2 diabetes, dyslipidemia, and hypertension, as well as there are some genetic factors that actually promote development and progression of the progressive form of NAFLD. Nevertheless, I think it's important to remember that the most common cause of death for patients with NAFLD is not liver disease, it's actually cardiovascular disease. A smaller proportion of patients with NAFLD that would predominantly have histologic NASH can progress, and they can progress to cirrhosis and its complications, including HCC. As we will share some of the data, one of the peculiarities of NASH is that patients can develop HCC in the presence of cirrhosis, but also in the absence of cirrhosis, patients with NASH can actually be at risk for development of HCC. The only allowed liver disease that you can see is that non-cirrhotic can develop HCC as hepatitis B. Now let's look at the data from the United States related to HCC and NASH. This is a model that we participated in a few years ago that estimated the changes in the incidence of NASH and then complications of NASH from 2015 to 2030. As you can see, HCC rates or incidence of HCC is actually also increasing. Of course, a little bit slower than decompensated liver disease and liver-related events, but it's going to be substantial as I'll show you some data in the next minute or so about its implication for liver deaths and liver transplantation. This is data from the United States looking at the causes of death from NRD, the registry that we have, and this is for 2017. We estimated about 73,000 liver deaths, and when you look at the causes of death here, about 17% were related to HCC. When you look at, again, this graph here, the bar graph, you see that hepatitis C and NAFLD were important drivers of this liver death in 2017. When you look at over time from 2007-2017, you see that hepatitis C is actually slowly declining at the top graph as the cause of cirrhosis in HCC. This happens to be for HCC, a similar graph you can also see for cirrhosis. On the other hand, NAFLD is actually increasing as a cause of death from HCC. In 2016, about 40,000 deaths in the United States were related to NAFLD. The bottom line is that when you look at the pie chart, that HCC was an important cause of death among NAFLD. When you look at the change over time from 2007-2016, the largest annual increase was for HCC, which was about 3.8%. Now let's look at liver transplantation. As you all know, NAFLD-related HCC is the second most common indication for liver transplantation for those who get listed for HCC, and I'll show you that data in a minute. This is data between 2006 and 2016 using the UNOS database for adult liver transplant candidates. About 160,000 candidates were here. About 16.5% of these were listed for liver transplantation. When you look at the HCC as an indication for liver transplantation in 2002, which was about 6.5%, it increased to about 23% by 2016. In fact, when you look at different causes of liver disease as an indication for HCC liver transplant, NASH was the most rapid. NASH-HCC showed the most rapid increase in being listed. In fact, between 2000 and 2017, about 12% of all NASH liver transplants were done for HCC. This is a little bit different than what I showed you before. This is looking at all NASH patients and say about 12% of them actually were done for HCC. And at the same time, when you look at all transplants of all HCCs, about 10% had NASH. Now, the interesting thing is that on the list mortality before getting transplant, the survival is the same. And when you look at post-transplant outcome of patients with NASH-HCC who are transplanted versus others, it's actually very similar. Now, this is what I showed you then when you look at liver transplant for indication that are not related to HCC. As you can see, hepatitis C is coming down and NASH is now going up, both alcoholic liver disease and non-alcoholic fatty liver disease is going up. When you look at the HCC liver transplant, again, the same data, the hepatitis C has declined rapidly for HCC liver transplant candidate. On the other hand, NAFLD is now second alcoholic liver disease as a cause of HCC listing for liver transplantation is actually the number one cause of being listed. Now, let's look at global burden of disease. And it's important to remember that liver cancer is the fifth and ninth most common cancer in men and women, accordingly. And it is the second most frequent cause of liver-related death globally. And of course, as you know, rates are higher in men. And of course, liver cancer includes other types of cancers. But hepatocellular carcinoma, which is really the topic of our webinar today, accounts for 90% or more of liver cancers. Globally, it's important to recognize that about 6 million incident cases of chronic liver disease was actually reported. About 16% of that was related to liver cancer and 84% related to cirrhosis. But when you look at deaths, liver deaths, about 2, 2.5 million liver deaths were reported. About 38% were related to liver cancer and 61% related to cirrhosis. So, the incidence is only about 15%, but liver deaths is actually almost twice as that. Again, there is an increase in liver complication in general. But when you look at globally, there are certain regions that actually show a decline in the burden of HCC in general. So, you look at China, it's in blue. And you look at the North American sort of area, it's actually suggesting there is an increase. The history of hepatitis B and how it's treated and managed in China maybe actually can contribute to the sort of declining profile. And the fact that we have naphthalene NASH quite prevalent in North America, that may actually explain what we are seeing here. Now, let me show you our own data. This is actually incident and mortality from HCC from different causes of liver disease using global burden of disease data. This happens to be for just HCC, different causes of liver disease. This is incidence and mortality, different regions of the world. I'm going to show you two things. Look at the hepatitis B in the green box. Most of these boxes are now in green, meaning that there is a decline in both incident and mortality related to HPV-related HCC. On the other hand, when you look at NASH, there is a lot of red boxes in terms of incidence and mortality. And that, of course, depicts the fact that there is an increase in both in incidence and mortality from NASH-related HCC. This is actually related to DALYs, which is not only mortality, now you're looking at disability issues. Again, there is a decline in HPV-related DALYs related to liver cancer, and there's an increase in NAFLD. So, the burden, the global burden of HCC in general is being now driven by non-alcoholic steatohepatitis. Now, one other issue that I'd like to get into, and this will be discussed later by my colleagues, is outcome of patients with NASH-related HCC. This is data that we published a few years back, looking at SEERS database between 2004 and 2009. This was a case control study looking at about 5,000 cases of HCC and 15,000 controls without HCC. And as you can expect, hepatitis C at the time was the most common cause of HCC, but so was alcoholic and non-alcoholic fatty liver disease. What's actually interesting is that when you look at NAFLD, the survival after one year of diagnosis is actually shorter for patients with NAFLD-related HCC. And this actually was actually significant in multivariate analysis. There are multiple reasons for this, and one may be that these patients don't get screened and they come in late. Another study that actually was a case control study, also smaller patient population, initially also suggested that NAFLD-HCC actually had a worse survival, and here, not HCV-related HCC, but once they controlled for confounder, the survival was no longer different. I think we're going to see this, that the fact that these patients present with larger tumors, more invasive of the vasculature, may be partly related to this, but other things, including the fact that they have obesity and type 2 diabetes, may also be drivers of this. Now, let's look at the incidence and risk factor. This is just to put this in the context for the United States. So, you can see that the HCC incidence in patients with, say, hepatitis C, is about 4%. So, the incidence in NASH is probably half of that, about 2%. But look at the total number of patients that you can actually could be at risk for, 9 million versus only somewhere about 2.5 to 3 million people. So, the numbers are large. Now, finally, getting into what are some of the risk factors for HCC. Some of these are very similar to what you've seen in terms of NASH. Metabolic risk factors, which are modifiable risks, are going to be risks. Cirrhosis, we know that presence of cirrhosis increases the risk. Male patients are at risk. This happens to be one of the fixed risk factors. Ethnicity and genetics. I'll review some of these with you. When you look at this, again, Cirrh's database over about, you know, maybe 12 years or so, and then, of course, there was projection. There was an increase, both in terms of male and female. And when you look at ethnicity, Hispanic males are at high risk for HCC. And this may again be driven by high prevalence of NASH in this population, as well as diabetes. And African American blacks were actually also at high risk. And this may have been driven primarily by hepatitis C. Now, genetics. I'm going to just cover this PNPLA3 genotype that's been associated with more progressive liver disease in patients with NAFLD. In fact, this is a population from the United Kingdom that they have looked at. And also some of the other European countries and looking at sort of their genotype. When you look at GG sort of allele, the G allele, each G allele that actually, you know, exists in these patients increases the risk of NAFLD-related HCC. So, the homozygous have the highest risk versus heterozygous, which was actually even higher than those who did not have any of these copies of this allele. So, there's some genetic predisposition. Now, the last issue here is a non-sorotic NAFLD. There are a number of case controls and case theories that suggested that. A very nice study came from Dr. Kahnwald's group and Dr. Hobserad's group from VA. 1,500 patients with HCC were looked at. A couple of things. About 13% of these patients were assumed not to have cirrhosis. And you can look at the data how it was actually generated. Level 1 was almost certainly did not have cirrhosis. And level 2 was high probability. And when you look at the different causes of liver disease, about a third of these of non-sorotic, of HCC patients and NAFLD were non-sorotics. And when you look at the risks of having HCC in the absence of cirrhosis, having the diagnosis of NAFLD and metabolic syndrome were associated with a non-sorotic HCC. So, what may be driving this? Well, again, the risk factor for NASH, diabetes, increases the risk of HCC by two to three-folds. A number of studies that are listed over here. I'm not going to belabor these. But just to show you a few studies. This is, again, from Dr. Hashim-El-Sorok looking at the veteran data and looking at those who have type 2 diabetes versus non-diabetics. There was an increase in terms of incidence of HCC. And this is another study that was published looking at a cohort of about 150,000 individuals who were discharged with diagnosis of type 2 diabetes. Now, you can see that the risk is actually increased in males specifically, in all patients, but specifically in males. Now, when you look at visceral obesity, and this is predominantly visceral obesity, that increases the risk of HCC by one-half to four-folds. Again, a number of studies that are listed over here, but I'll just share a few of them with you. This first study was from a transplant analysis that was published almost two decades ago. But when you look at, again, BMI of more than 30 and diabetes, they were actually independently associated with incidence of HCC. Another very interesting study from Denmark. This is a Danish cohort of almost 290,000 children. It really depended on what the children's weight was when they were younger and what happened to the incidence of HCC later on during life. So, liver cancer, the hazard ratio for adult liver cancer was 1.2 and 1.3 for each unit of BMI at age 7 and 13. So, younger age obesity will actually increase the risk of HCC later on. This is a case-controlled study of young adults. And again, similar data that in the adulthood, the risk increases substantially. In fact, if you have type 2 diabetes and obesity, the risk is even higher, as you can see in the bar graph. So, in summary, HCC is an important complication of non-alcoholic fatty liver disease. In the United States, NASH HCC is the most rapidly growing HCC indication for liver transplantation. HCC can be promoted by metabolic abnormality that are closely tied with NASH, the insulin-resistant type 2 diabetes, and visceral obesity. Given the very high prevalence of Nalphaldehyde in NASH in the general population, and also the fact that you can get HCC in the absence of cirrhosis, the number of patients with HCC due to Nalphaldehyde are expected to grow. They are already sort of becoming a major part of our liver clinics' referrals. Some cases of Nalphaldehyde, as I suggested, is going to be seen in non-cirrhotic. Of course, the dilemma here is how to screen for these patients because they currently don't meet the screening threshold for HCC, but there's got to be a better way to actually manage these patients. And lack of effective screening for NASH right now and other factors really can lead to worse prognosis for NAFLD-related HCC. I'm going to stop here, and again, thank you again for listening to this presentation, and I'll be glad to take questions at the end of this session. Thank you. Thank you very much, Dr. Yunossi, for an excellent talk. For the attendees, please add your questions that you have for Dr. Yunossi and his talk in the Q&A box. It's a great pleasure to introduce our next speaker, Dr. Julia Vita-Cheryl. She will review the clinical manifestations of NAFLD-related HCC with us. Dr. Vita-Cheryl is an Associate Professor of Medicine and Director of Non-Alcoholic Fatty Liver Disease Program at Columbia University. Her clinical and translational work spans the multidisciplinary care of NAFLD patients at all stages of their disease, but her specific focus is on precision medicine and investigation of rare genetic variants that influence the development and progression of NAFLD both before and after liver transplantation. She's also done work around risk factors for HCC, and she is here with us today. Thank you very much, Julia, and looking forward to your presentation. All right. So, my charge today is to talk about the clinical manifestations of NASH and HCC as they coexist in patients. For the purposes of consistency with the other panelists today, and given some of the limitations that we know about the histologic confirmation of the disease, I'll be using the terms NASH and NAFLD-related HCC interchangeably. These are my disclosures. So, given this audience and expertise in the diagnosis of HCC and the diagnosis of fatty liver disease, one of the objectives is to increase our incremental knowledge, and how do we do that? So, given what we've seen about the emerging evidence within the fatty liver disease population, can we identify some key clinical manifestations that's associated with HCC in individuals without cirrhosis? So, increasing our clinical acumen, just as Dr. Yonasi just mentioned in terms of screening and surveillance of these individuals. How do we identify epidemiologic outliers, and can we name specific differences that occur in unique subpopulations who develop NAFLD and NASH-associated HCC, and can we describe some distinct histologic features that are associated with NASH-HCC in the context of knowing that most of these patients are diagnosed radiographically? What can we learn about their histology that will inform some of the tumor-directed therapies? So, clinical features of patients with HCC that arise in NASH and NAFLD, as has been mentioned before, Dr. Conwall's impressive study that helped us glance at the NAFLD population versus controls, so individuals with chronic liver disease that was not classified as NAFLD. A couple of particular patient population, this is a VA patient population, so predominantly male. NAFLD was used in the absence of hepatic steatosis. This was biochemical analyses, so patients that had biochemical elevations in their aminotransferases, and the diagnosis of cirrhosis that we will get to was predominantly ICD-9 codes and complications of cirrhosis. So, when we juxtapose a NAFLD group with those who do not have fatty liver disease as their chronic liver disease, what are some clinical observations that we can make about the patients that bear these potential risks? Again, age and sex-matched individuals. Comorbidities were obesity, overweight status, more in individuals with fatty liver disease, more individuals with fatty liver disease had cardiometabolic comorbidities like diabetes and hypertension, very consistent with some of the global observations that we've just heard about, and in individuals that do have cirrhosis, more of them were reflected in the fatty liver disease group as opposed to those that had controls. So, this is the population that Dr. Yonasi had mentioned in terms of our epidemiologic studies, but how can we make those incremental gains as clinicians in terms of clinical observations and manifestations in individuals who don't bear these characteristics? Part of that is informed by another study that we've given brief mention to prior, and that's the MUTAL study, also a VA population, also predominantly male. What did we observe in those individuals that did not have cirrhosis by level one or level two criteria and had a diagnosis of fatty liver disease? These individuals typically were older, their median meld was lower, as would be expected. These included cardiovascular comorbidities, so type two diabetes, cardiometabolic diseases reflected by the incidence of myocardial infarction and peripheral vascular disease. Biochemically, these individuals, as would be surmised, had a lower AFP, and that may hint at biological differences, tumor biology differences, as opposed to constitutional biology differences that were extant in this population as well. And then their BCLC stage, more individuals presented with stage B, as opposed to stages C and D. Some of those fit with expectations, and others are a bit of a surprise, and again, increasing our clinical index of suspicion for individuals that may present atypically. So zooming out from that population in individuals that have HCC that arises outside of a diagnosis of cirrhosis who don't fall into regular screening and surveillance mechanisms, what have we learned? So this is a population that both Dr. Garrier, who you'll hear about, who you'll hear from next, and I participated in. It's a longitudinal cohort that looked at five centers that specifically wanted to try to address some of these questions. It was over a pre-direct anti-acting antiviral stage in the history of our chronic liver disease spectrum. So between January 2000 and 2014, five academic medical centers, including a cancer center, as we see some observations and some diagnostic and treatment modalities are different based on center. This is an adult population only, and specific to the 90% of patients that have a diagnosis of classical HCC as opposed to some of the other sub-phenotypes. And it was over 5,000 patients that were included that allowed us to drill deeper into understanding some of the outlying features in individuals that developed HCC independent of cirrhosis. So there's additional evidence here of the different etiologies of chronic liver disease. This is also classified using the Mital criteria, so level one and level two, if I failed to mention that earlier. So in individuals that developed HCC, the bulk of the individuals that had a diagnosis that was attributable was in individuals that had non-alcoholic fatty liver disease. Of note, as opposed to the Conwell study, which we described earlier, hepatic steatosis was included in this group as part of the diagnostic criteria. You'll see also that a large percentage of patients fell into an unknown category and couldn't be classified by expert classification based on available data. So a lot to be done for future longitudinal cohorts and registries in terms of classification prospectively. So what are some of the key take-home clinical features that we as clinicians can start to observe and classify as we see patients and then quantify in our electronic health records to make some of these studies in the future more possible and able to make some linkages in our associations more strongly? So the comorbidities, many of them are expected. Higher BMI, the presence of obesity, hypertension, dyslipidemia, coronary disease, peripheral vascular disease. The history of alcohol abuse is one that's notoriously difficult to capture in electronic health records and quantify on manual chart review. But that is an aspect that can interplay with the development of the metabolic milieu that allows for metabolic reprogramming that we observe in HCC that develops outside of cirrhosis. And then some of the biochemical parameters that we're getting better at capturing. So the total bilirubin, INR, MELD score, the AST to platelet ratio index that's classified, and the FIB4 categories that are associated with more advanced fibrosis, higher risk of development of premalignant and malignant features. Now, given the multiple parameters and the multiple modalities that we view the average human who comes to us seeking care, what are some of the benchmark characteristics that we have observed in individuals who have developed HCC in the setting of fatty liver disease? So again, in the absence of cirrhosis, so tumor characteristics, as has been mentioned, in addition to what we know about the general population, the cirrhotic population where they present typically with larger tumors and more advanced disease, these individuals as well, whether or not it is a failure in detection or underlying biology or something that's not captured yet in the data set, tumor size and AFB are distinguishing characteristics. The surveillance modality, which our next speaker will also talk about, whether or not it's regular surveillance versus the incidental capture and then catch-up care, as we'll see in some of the ethnic distribution differences, staging features, both anatomic stage, most of these individuals did present within Milan criteria and their tumor staging. And then lastly, histology, which is a much smaller subgroup given the absence of histologic diagnosis of HCC in many patients that are not captured through surgical resection or transplantation. So what are our take-home messages? NAFLD is the most common etiology in patients without cirrhosis, and that NASH-HCC can present with more advanced disease and a larger tumor burden. Whether or not that impacts survival is a subject to, again, larger longitudinal registries and in different subpopulations, we have found that that not to be the case. So as opposed to biological sex, self-identified gender, when we looked at the data set to look at some of these subpopulations that we may be missing in some of the larger pool data sets that are obtained from different distributions of diverse cohorts, the etiology of non-cirrhotic NASH-associated HCC tends to be very common in women, NAFLD tends to be common in women as the leading etiology. As you can see, hepatitis C alone in this data set was also a major player, but what did we observe? Again, incremental clinical knowledge in a female population as opposed to male is that women presented at an older age, they less often had cardiac disease, they often presented with a lower MELD score, and some of the biochemical parameters that were used to capturing an electronic health record reviews, AST and ALT, were also lower in women. In patients that present without cirrhosis, we see a couple of other observations. So more commonly, women present without cirrhosis, that's one take-home message, and they also present with a single tumor as opposed to multiple tumors. Again, underlying biology within the liver, within the cancer itself, or unmeasured postmenopausal, all limitations of the data set here, but a clinical observation. So, very consistent with Dr. Connell's study that looked at some of the advanced age and Hispanic ethnicity components to individuals that presented with NAFLD-HCC in a cirrhotic population, what did we observe in our registry? So tumor characteristics and underlying etiology demonstrated that Hispanic patients had a high proportion of non-alcoholic fatty liver disease as a contributing etiology. We know about the PNPLA3 phenotype, both associated with alcohol and non-alcoholic genotype, associated with alcoholic and non-alcoholic fatty liver disease, and that is thought to be one potential component to the underlying observations, in addition to access to care, in terms of access to presentation, presenting symptoms, and surveillance and screening mechanisms. Patients presented with more advanced disease at presentation, and so that may reflect the social determinants of health as opposed to the biological determinants of health. Again, more preliminary data for our longitudinal cohorts going forward. Now, when it comes to potential targeted therapies, where many of us are working within the fatty liver disease world versus the cancer world, what have we learned based on some of the biological characteristics of classical HCC, and how might those tumor characteristics play into some of the treatment modalities that Dr. Gaurier will speak to next? About 10, 12 years ago, Marcella Saladino here at Columbia published some of the early work on the steatohepatitic variant of HCC, and these were the observations that were noted. There was a loss of cytoplasmic CK8 in 18. There were increased numbers of activated hepatic stellate cells with this steatohepatitic hepatocellular pattern. We'll see it later described as the non-proliferative type. It also bears a pattern identical to that seen in steatohepatitis in non-neoplastic liver. ER stress, pathologic lipophagy, and increased reactive oxygen species production and the reducing power could cause some of the, these are all potential mechanisms by which the oncogenic genetic alterations that were seen in overloaded hepatocytes, overloaded by fat, that may promote some of the malignant transformation. Again, some of the metabolic reprogramming that's been postulated as a potential mechanism. And all of this portends molecular signatures that are unique to the NASH and NAFLD intrahepatic, multifocal, heterogeneic type of milia that these cancers are arising in. So with that, I will leave some of the directed therapies to Dr. Gaurier next. And in summary, obesity and cardiovascular comorbidities are more common in NASH HCC, very consistent with what we've seen in the global disease burden and the trajectories of fatty liver disease as a main contributor to chronic liver disease mortality overall. Women are more likely than men to develop HCC in the absence of cirrhosis and the molecular phenotyping of tumors are likely to reveal some biological contributors to these clinical observations across gender and ancestry differences. Thank you. Thank you very much, Dr. Gaurier, for a brilliant talk. Our next speaker is Dr. Samer Gaurier, who is an Associate Professor of Medicine and Director of Hepatology Research Fellowship Program in the Division of Gastroenterology and Hepatology at Indiana University School of Medicine. His clinical and translational research focuses on different aspects of non-alcoholic and alcoholic fatty liver disease, including machine learning-based histological phenotyping. He's also interested and has done work in genetic modifications and diagnostics. He studies HCC associated with these conditions with a particular focus in non-cirrhotic HCC. It's a great pleasure to have you here, Samer. I'm looking forward to your presentation. Thank you so much, Fasiha. I'd also like to thank the NASH-SIG for the invitation. I'd also like to thank Dr. Yannossi and Dr. Botticello for the nice setup for my talk, and thank you for all attendees. So, with that, let me start my talk. So, these are my disclosures, and this is the outline of my talk. So, starting with surveillance, the current ASLD guidelines recommend that HCC surveillance be given to adults with cirrhosis, with ultrasound, with or without alpha-fetoprotein level every six months. The aim of surveillance is to improve early detection and the receipt of curative therapy. So, based on this modality, which is driven, these modalities, the data are driven from mainly observational studies and have low quality of evidence. This is just to keep in the back of our mind. One challenge with HCC surveillance is that there's poor adherence to it across the board in patients with cirrhosis, and particularly in patients who have NASH cirrhosis. The pooled adherence to surveillance in a recent meta-analysis was a dismal 24%. Now, that improved to about 74% in subspecialty care studies. Now, the reason adherence to surveillance in such a poor state may be related to different barriers. Some of them are patient-related and some of them physician-related. Addressing these barriers may indeed improve adherence to the surveillance. Now, it is important when someone applies surveillance for HCC that the attention to the quality of ultrasound receives, you know, receives the adequate attention, basically. Is the ultrasound adequate or not? We know from different studies that the higher the BMI, the higher the failure rates of ultrasound. And, in fact, if someone has a body mass index over 35 or have a NASH cirrhosis diagnosis, the failure rate of ultrasound could be over a third of the time. So, that's really important to take account of. Now, we've learned recently that patients with NAFLD at highest risk for developing HCC are actually those with cirrhosis. Whether they have high FIP4 score or not, they are at the highest incident, they have the highest incident rate of HCC development, whereas patients with NAFL without cirrhosis are at much lower risk. This important study yields support to the current ASLD guidelines that recommend that surveillance is offered to patients with HCC, and a recent panel of experts suggested that it should be extended to NAFL patients with advanced fibrosis based on non-invasive testing. Now, as I said earlier, the quality of ultrasound should be documented, and if it is inadequate in a particular patient, that patient may be given the opportunity to have screening or surveillance with CP or MRI with or without AFP every six months. The new trend in the field is to risk stratify the patient for HCC in the setting of NAFL. Here you see in the left panel a study looking at polygenic risk score, and as you can see, the higher the number of risk alleles from candidate genes that influence liver disease, the higher the risk of the patient for developing cirrhosis or HCC, and they validated their findings in other cohort. This is nice, but there is even more practical model which uses available metabolic traits and, you know, that are readily available, and you could see that patients who have more than one trait, the risk goes up, and the higher the number of the metabolic trait a patient has, the higher the risk for developing cirrhosis and HCC in the setting of NAFL. So, these models are very exciting, but require further validation before implementation and practice. Moving on to prevention, there is plenty of cohort and case control studies that suggest that coffee consumption may reduce the risk of HCC, and this was the case even with decaffeinated coffee, but to a lower extent than caffeinated coffee, and you can see that there is a dose effect. The higher the number of coffee drinks consumed, the more the risk reduction. There is data from the surgical literature that suggests bariatric surgery of any type of bariatric surgery, whether it be a true-on-Y or sleeve gastrectomy or adjustable band, could actually reduce the risk of HCC. The problem with this literature is that there is considerable heterogeneity in the studies looking at this, but nonetheless, it makes a plausible biological sense that improving obesity may reduce the risk of HCC, as Dr. Yannosi and Dr. Wathicharel alluded to. Metformin actually may reduce the risk of HCC in patients with type 2 diabetes. This is an example based on the Taiwanese health registry, where patients with type 2 diabetes and across different liver diseases showed reduction in the risk of HCC with metformin use. At Indiana University, we have a well-phenotyped NAFL cohort with biopsy that has advanced fibrosis, and in our cohort, we found the same thing, that actually metformin users run a lower risk for developing HCC than non-metformin users. A recent Swedish study showed that using aspirin at low dose is actually associated with low risk of developing HCC and liver-related mortality in patients with viral hepatitis. This came at no significant increase in the 10-year risk of GI bleeding. Now, whether these exciting findings apply to the risk of NAFL HCC remains to be studied. Using our cohort at IU that, again, is phenotyped with biopsies and has advanced fibrosis, we looked at the effect of vitamin E on heart outcomes, and we noted that vitamin E improved transplant-free survival and hepatic events in these patients, but it did not improve the risk of HCC. Lastly, on the prevention end, there was a lot of excitement generated by an earlier meta-analysis and echoed here in this updated meta-analysis that statins may reduce the risk of HCC both in type 2 diabetes as well as in cirrhosis. However, the data from the three available RCTs or meta-analyses of these RCT data show wide confidence interval that crosses the zero and do not yield support to the findings of the observational case control and cohort studies that statins reduce the risk of HCC in this population. Moving on to the treatment approach, as you know, the Barcelona Clinic liver cancer staging system is widely accepted by different societies, and it is based on tumor burden, liver function, and performance status. Other critical factors in decision-making include the presence of clinically significant portal hypertension, the cirrhotic status of the patient, and comorbidity. Ultimately, designing an effective treatment strategy is a multidisciplinary sport that involves a tumor board review and multiple disciplines to take care of the HCC as well as the involvement of disciplines that help manage NAFLD, such as bariatric surgery, nutrition endocrinology, and other specialties that help manage the considerable burden of comorbidities in these patients. So, next, we're going to look at the data for managing different, treating different stages of BCLC-HCC, and we're going to start with data for very early and early-stage HCC. These are the patients that should receive curative treatments. So, starting off with ablative treatments, the data suggests that radiofrequency ablation suggests that radiofrequency ablation offers the best survival of ablative tools. It's about 70% at 5 years. Other ablative tools are good and effective, but to a lesser extent. The question whether adjuvant therapy following resection or ablation is useful has been answered as no, as data from seraphinib plus sirolimus-based immunosuppression studies showed no effect on survival. Importantly, the ASLD guidelines favor resection over RFA for early-stage HCC because mainly the RFA data is driven from Asian populations with viral hepatitis B. Now, if you have a patient who is a potentially surgical candidate, the key factors that are to be considered in determining the risk of resection are the presence of clinically significant portal hypertension, the extent of hepatectomy, is it minor, three hepatic segments or less, or major, and then the MELD score. Now, patients who fall in the high-risk category should be offered different modalities because there is really very high risk of liver decompensation reaching 60% and liver-related mortality can be up to 25%. So, who are the ideal resection candidates? Well, those without cirrhosis with resectable HCC or those with cirrhosis and resectable HCC who have compensated cirrhosis, child's A, without the clinically significant portal hypertension. Now, if a subject is offered or if a patient is offered resection, ideally the resection should be performed in an atomic fashion following the hepatic segments anatomy to reduce the risk of tumor recurrence. Now, the limited data available on outcomes of surgery in these patients suggest that they do just fine. In this study from Italy, the outcomes of metabolic syndrome associated HCC were compared to the outcomes of HCV-related HCC, and as you could see in terms of all complications and morbidity, they were similar, no differences, and indeed patients with metabolic syndrome-related HCC had slightly better survival than those with hepatitis C-related HCC. For patients who are listed for liver transplant for NAFLD-HCC, as you could see from this important study from Dr. Yonasi, that their transplantation rate is not different from HCC related to other liver diseases, nor is their dropout from the list rate different from HCC related to other liver diseases. Importantly, their post-transplant outcomes is similar on and in some parameters better than HCC from other liver diseases. The European Transplant Registry provided further support to these findings that patients with NASH-HCC have similar patient and graft survival to HCC related to other liver diseases. How about optimizing the patients with sleeve gastrectomy at the time of liver transplantation? This is a provocative approach, but makes sense, and this is the result of a NICE case series from Mayo Clinic Rochester, in which you see there were a few patients who had NAFLD-HCC that started off with BMI over 40 and did very well with the combined operation. In general, you could see the data here for post-operative outcomes. Those with combined liver transplant and sleeve gastrectomy had much more dramatic weight loss that was sustainable post-transplant compared to those who underwent only medical weight loss, and overall the frequency of metabolic complication was much lower in those who also underwent the sleeve at the same time. How about the bridging therapy? This is widely used. There is really weak evidence to support the use of bridging therapy, as you see here in this meta-analysis. Nonetheless, there is a trend for improved dropout rate from the waiting list, as well as improving mortality and survival on the waiting list. Now, the ASLD does not favor one local regional therapy over the other to achieve that goal. Moving on to intermediate-stage HCC, here we're going to look at the data for transarterial therapy, which are the main modality of treatment. The best survival is actually for TACE, whether it be conventional or DEP TACE, where you get a median overall survival of 30 months. Now, adding kinase inhibitors to it did not really improve survival, and then when you look at transarterial embolization, commonly utilized also in practice, the initial promising data, especially with vascular invasion and portal vein thrombosis, did not pan out with subsequent study, and the available evidence suggests that the overall survival may be less than that of TACE, but there are no head-to-head comparisons. Next, we're going to talk about the evolution of systemic therapy for advanced-stage HCC. This has witnessed a proliferation of data and the availability of second- and third-line options for patients with advanced HCC. So, it's been 13 years since the publication of the landmark SHARP study, which established the superiority of sorafenib over placebo, offering a median survival advantage of about three months. Since then, it took over a decade to find a superior regimen, which is the combination of etizolizumab and bevacizumab over sorafenib. You could see they offered much better survival. Now, here you see the listing and mechanisms of approved agents for advanced HCC, which include a combination of immune checkpoint inhibitors, monoclonal antibodies, as well as kinase inhibitors. Those are approved based on Phase II and Phase III studies in the United States. Now, we have looked at the utilization of different modalities for HCC together with my colleague, Dr. Wattachero, and as you see, when you look at the utilization for cirrhotic status, the most common modality for treating HCC in patients without cirrhosis is actually resection, followed by catheter-based treatment and sorafenib. Now, when you look at cirrhotic patients, you will notice that catheter-based treatments are the most commonly utilized, followed by transplant and sorafenib. Now, when we looked at the data specifically for NAFLD per cirrhotic stages, we saw similar trends of utilization to the overall NAFLD in this cohort of over 5,000 patients. So, to summarize, surveillance is currently recommended to be performed by using ultrasound with or without alpha-fetoprotein every six months. Currently, adherence to surveillance is poor, and we have little emphasis on prevention in practice. There is an evolving role for chemotherapy in managing advanced stages of HCC. The outcomes of NAFLD-HCC is good and similar to HCC from other etiologies. Future studies obviously need to optimize surveillance for NAFLD-HCC with the specific challenges encountered, including improving adherence, optimization of the modalities and intervals. We need validated risk stratification models to improve the efficiency of surveillance. The role of improving risk factors, such as obesity or things that reduce the risk, such as coffee and aspirin, in NAFLD-HCC risk reduction needs to be further studied. Certainly, with the exciting data about sleeve gastrectomy outcomes, this has to be studied further, and its role in optimizing surgical candidates definitely is important. Finally, the role of novel biomarkers and tumor-liquid biopsy for diagnosis, screening, as well as assessing treatment response need to be further studied. Thank you for your attention. Thank you so much, Samer. Excellent overview and overall great session altogether. We have a few minutes, and we have a few questions that I can go over and then see if there are any other questions that come up between now and then. So let me start with a question, and Dr. Zunigunose, if you can take this one. It's from Dr. Jenny Yang. It is extremely challenging using current clinical-approved MRI contrast agents to detect fibrosis or cirrhosis, especially with the background of fatty liver in the liver. So the question or comment is that it's quite possible that NAFLD as a major contribution for non-cirrhotic HCC is partly due to the challenges or the incapability of assessing stage by the current imaging modalities. What is your thought on that? No, I think that is absolutely correct. That may be part of the confounders and problems that we have, but if you look at even histologic data where you have actually to see resected in some of the case series, and you look at both the tumor as well as the surrounding tissue, there are a number of cases that will have not really significant fibrosis of the liver and the non-HCC part of the liver. So I think there is no doubt that there may be some misclassification, but also there is no doubt that HCC can occur in the absence of cirrhosis. The exact incidence can be debated. Maybe it's underestimated or overestimated, but my own feeling is that that's where the genetic, the metabolic risk factor, especially diabetes and obesity play a much prominent role than we would have otherwise considered. The one other thing that I would suggest is that it's not just fatty liver disease that obesity and type 2 diabetes actually promotes HCC. If you look at, for example, presence of type 2 diabetes and obesity in patients with alcoholic liver disease or viral hepatitis, the risk increases, and I think that is sort of the pathogenic pathway that I can at least envision contributing to the development of non-cirrhotic HCC. Thank you, Nakunose. The next question, if Dr. Bhattacharyya, if you could take this question. Some papers show that some hormones like estrogen, they prevent lipid from being accumulated in the liver of a woman. Is there any difference in the pathogenesis of cirrhosis, I believe by sex and gender? Yeah, great observation. This gets at some of the limitations of using in vivo data with clinical populations. So the biological availability of estrogen to that liver and its contribution to cirrhosis rather than a snapshot in time, a cross-sectional assay would be the way to do this. So yes, estrogen is thought to be very hepatoprotective. We use that kind of general language. But the role to which lipid accumulation plays a role in insulin resistance, liver fibrosis, and then risk for cancer. I think this question gets more at the risk of cirrhosis than potentially the risk of cancer. The two populations that we've studied probably the most are fatty liver disease and autoimmune hepatitis, female predominant or more clinically observed. But to my knowledge, we haven't looked at serial assays over a longitudinal period of time to look at the direct contribution of estrogen to the liver, to the metabolism of what's going on. And most of the data that we have relates to cirrhosis and the observation of estrogen in both men and women after the diagnosis of cirrhosis. So more to be done, more to be studied. Thank you. Thank you very much. And one other question, and if you could take this one, are there any data on patient characteristics that are associated with poor adherence to surveillance and how that might relate to potentially worse outcomes in NAFLD, HCC, especially in minority groups? So as I shared earlier, adherence is overall poor. It improves considerably deliverance of surveillance in subspecialty care. Up to 74% of patients who attend specialty care with liver disease get surveillance. And Dr. Singal from, you know, Texas had done actually important work looking that minorities, yes, are disadvantaged. There are many barriers that they face from transportation, to knowledge about the disease process, to access to care. All these are real barriers that patients experience. Now, addressing these barriers, as I said, has the potential of improving adherence to surveillance. Thank you. Thank you very much. I think we are right at the top of the hour, maybe a few minutes over. It was excellent overall. We know you guys are all excellent speakers and you kept up the tradition. Brilliant talks. And for all the attendees who joined us, thank you very much for joining us. If you have any other questions that come up, please feel free to send them to us and we will pass them along to all the speakers. Thank you again. Good to see everyone. Take care. Thank you very much. Bye-bye. Bye.

AASLD webinar

hepatocellular carcinoma

HCC

NAFLD

NASH

surveillance

cirrhosis

barriers to adherence

risk stratification models

treatment strategies