Good afternoon, and welcome to today's webinar on genetics and NAFLD. I'm Anna Medell from Duke University in the U.S. We have an exciting program with three experts who will update us about how genetic variation impacts on NAFLD susceptibility and progression. Our first speaker is Dr. Sylvia Estukyan. She's from the Institute of Medical Research in Buenos Aires in Argentina, and she will provide an overview of genetic variants that have been associated with NAFLD. Our next speaker is Dr. Yaron Rotman from NIDDK. Yaron will tell us more specifically about how some of the variants influence particular aspects of NAFLD pathobiology. And our final speaker is Dr. Luca Valente from Milan, Italy. Dr. Valente will give us a more in-depth discussion about this issue and tell us how some of the gene variants interact to influence NAFLD progression, providing the basis for efforts that will hopefully eventually lead to the development of combined genetic risk scores that can help us manage NAFLD patients in the clinic. And we're going to move from one talk to another without interruption and address the questions that you might have at the end of all of the presentation, so please submit your questions via the Q&A box, which is at the bottom of your screen, and then I'll sort them and direct them to the appropriate panelists. We are not taking oral questions, obviously, during the webinar, but we will be responding to your questions verbally in a panel discussion at the end of the webinar. So without further ado, Sylvia, take it away. Okay, thank you, Anna-Mae. Okay, hello, everyone. Welcome to the Genetics of NAFLD webinar organized by the SLD Genetics Non-Alcoholic Fatty Liver Disease Interest Group. Today, I will illustrate the basis of genetics in NAFLD. This is my disclosure. We all know that NAFLD has emerged as the leading global cause of chronic liver damage. More importantly, NAFLD presents phenotypic complexity, meaning that the disease can progress into severe clinical forms, including NASH fibrosis, cirrhosis, and even hepatocellular carcinoma, and inter-individual variability, meaning that about 30% of the patients with fatty liver will progress to NASH, and about 20% of NASH patients might develop cirrhosis. Then, I will show you that genetic variation underlies differences in disease pathogens. Before starting, I will introduce two key concepts for you, that is the definition of a monogenic and polygenic disease. A monogenic disease is associated with a phenotype produced by an alteration of a single pathway. Monogenic disease is run in families and can be dominant or recessive and autosomal or sex-linked. On the contrary, a polygenic disease is caused by the combining effect of more than one gene and multiple pathways. Thus, we can say that NAFLD is a polygenic complex disease, is heritable, is influenced by the environmental exposure of lifestyle, and co-occurs with one or more cardiometabolic risk factors. Let's start with the premise that NAFLD is a polygenic and heritable threat. Keystone studies, including familial aggregation and twin studies, candidate genes and genome-wide association studies, and experimental evidence, show that demonstrated that NAFLD is under the influence of the effect of several genes. NAFLD heritability estimate is about 20 to 50 percent, depending on the study design, ethnicity, and the methodology used to characterize the phenotype, which is quite similar to the heritability of other complex traits. For example, the heritability of gastrointestinal diseases is around 28 percent. The majority of genetic variants are tolerated by an organism, but some variants can change the function of a gene product. So, it is important to understand the molecular mechanism and pathway that link genotype to phenotype. Deciphering how associated variants modulate the disease risk and severity, and how they impact cellular phenotypes, and enforce mechanistic therapeutic hypotheses, resulting in more drug discovery and, of course, personalized means. For example, when SNPs are located near a gene, then it is postulated that the gene contributes to the variation under investigation. Genome-wide association causal variants have been proposed to influence the disease risk by altering the function of cell type-specific regulatory elements. Testing biological function of a variant can be done in cells or animal models. eQTL, expression-quantitative trait loci, is a locus that explains a fraction of the genetic variants of a gene expression phenotype. And some SNPs are known to interact with early life events to affect the risk. Natural selection results in DNA sequence differences among individuals, populations, and species. So, let's move to the genetic variants associated with natural susceptibility. Discoveries of variants of genes that regulate metabolic traits represent a large proportion of the genetic components of natural and natural are shown in this cartoon. Here, I show you that targets associated with a fatty liver are based on importance and novelty. The plot depicts targets associated with natural with emphasis on proteins linked to lipid metabolism. Targets are plot with the log-log importance and novel axis according to the algorithm implemented by the web tool resource that explores the relationship between proteins and diseases extracted from the scientific literature and which is named GeneX resource. You can see here that targets associated with strong and likely a gradient number of publications are located in the upper left part of the plot. Most relevant and replicated targets are illustrated in this figure for you that show the most relevant protein function as well as cellular localization. Starting with PMPL3, it's a triglycerol-like path that mediates triglycerol hydrolysis and adipozytes. It is also a metabolic intermediate and precursor of both triglycerides and glycerophospholipids and is mostly located in lipid droplets. TM6XF2, which is a regulator of liver fat metabolism, influencing triglyceride secretion and hepatic lipid content, and also HSD17V13, which is a hepatic lipid droplet protein with oxidoreductase activity. MVO87 is another target that participates in the recylation step of a phospholipid remodeling pathway, also known as the LAM cycle. And finally, GCKR is a regulatory protein that inhibits glucocortisone in the liver and pancreatic IL cells that is located in the nucleus, mitochondria, and cytoplasm. You can see here that HSD17V13, TM6XF2, and MVO87 are all located in lipid droplets, endoplasm, reticulum, Golgi, intermediate compartment, and sometimes in the membranes. It is important to remind you that protein localization in subcellular compartments is of paramount importance for the processes of intracellular communication and also cell signaling. This BC table resumes for you the associated variants with the risk and with the protection of natural disease. You can see in green letter the protection variant, and in the black, all the variants that confer risk. This BC table resumes the variant ID, the most severe consequences, the global minor allele frequency, the functionality, the effect, and also some key references. As you can see here, the non-synonymous variants in BMPL3 is regarded as the major genetic component of NAFL and NASH. The risk effect of this variant on developing fatty liver is the strongest ever reported for a common variant modifying the genetic susceptibility of NAFL. It's about 5% of a variant. This plot, based on UK BioVamp population, shows you that the variants in BMPL3 is probably the major determinant of other chronic liver diseases, including alcoholic liver disease, fibrosis, and cirrhosis. Surprisingly, it seems that the variant confers protection against cholelithiasis and gallbladder-associated diseases. Knowledge integration and data modeling and analysis are vital processes at the interface with drug discovery. Then let's move to the genetic pathways in NAFL. The major candidate gene variants function in metabolic pathways. This cartoon illustrates that variants associated with the genetic susceptibility of NAFL are linked to substrate delivery for the novel lipogenesis, mitochondrial energy utilization, lipid droplet assembly, lipolytic catabolism, and fatty acid compartmentalization, as well as BLE assembly separation. Also, we know that we gain a lot of knowledge about BMPL3, and this figure shows you the master network in which BMPL3 is involved. A large proportion of the pathways are linked to lipid metabolism, from lipid droplet organization to adipogenesis, suggesting that BMPL3 is involved in the pathway of transcriptional and hormonal regulation of adipocyte formation. We can say that variants associated with the risk of NAFL and NASH are common variants associated with modest effects. For example, you can see here the effect of BMPL3 and the effect of HD17B13 that confer risk of protection against NAFL and NASH with a significantly moderate effect. These are followed by the modest effects of TMC-SCF2 variant, also the variant in GCKL and in BOT787, which have a smaller effect on the disease risk. But it remains poorly explored the role of rare variants, which probably would have substantial effects in the phenotype. So, a significant proportion of the disease burden could be explained by the misinheritability, which covers not only genetic and epigenetic modifiers, but the interaction with environmental exposure, as well as with a highly interconnected and dynamic network of pathways. Some of them are illustrated in this figure. Mapping the genetic component of NAFL and NASH should include not only the search for rare variants, but the exploration of structural variation, for example, copy number variants. Also, given the role of mitochondria in the physiology of the disease, it is also worldwide to characterize the genetic diversity of the mitochondria in AIR. And finally, it is plausible to presume that NAFL and NASH visibility gap might be explained by the intricate relationship among genetic variants of the nuclear and mitochondrial genome, the phenotype, and the yet unexplored interaction with epigenetics and environmental factors, including the microbiome. So, to illustrate the topic of the misinheritability of NAFL, I will share with you a clinical case. This is the case of a 48-year-old male who, by the time of first clinical consultation, present with elevated level of aminotransferases, arterial hypertension, morbid obesity, and type 2 diabetes. Liver biopsy show NASH without liver fibrosis. The patient gained weight and presented persistently elevated aminotransferases. The second biopsy, five years apart from the initial one, showed that the patient developed NASH fibrosis. We found that the initial, the patient had a rare non-cell mutation in the glucokinase regulatory chain that caused decreased protein expression. Thank you for your attention, and we would be happy to take questions at the end of the webinar. Thank you, Dr. Sukhoian, for a fantastic presentation. I'm Yaron Rotman from the NIDDK, and I will be discussing about the role of specific variants in humans with NASH. These are my disclosures. So, I will discuss the specific variants in these four genes. There are, as Dr. Sukhoian mentioned, there are multiple genes that are associated with NAFLD, but I'll focus on these four. And I will focus on their association with specific phenotypes with the amount of liver fat, with the presence of NASH, fibrosis, hepatocellular carcinoma, and mortality. And at the end, I'll mention a little bit about the combination, combining multiple genes to a combined genetic risk. So, if we use this very simplistic scheme of the pathogenesis of NAFLD and NASH, we think there's a combination of, there's a overload of caloric intake that leads to accumulation of liver fat. And in some subjects, we are developing metabolic and oxidative stress, hepatocyte injury, inflammation, and development of fibrosis. And this correlates roughly to the histological transition from steatosis to NASH and to cirrhosis. And theoretically, we could have genetic variants that affect each and every of these pathways. We could foresee a genetic variant that only affects fibrosis, or that only affects the inflammatory process, or a genetic variant that affects accumulation of fat. And we should remember this for the end of the talk. The first variant that was identified as a strong link with NAFLD is PNPLA3, the patatine-like phospholipase domain containing three. This is likely the most commonly shown slide in talks about NAFLD genetics. In two genome-wide association studies, one from the Dallas Heart Study focusing on association with liver fat, and another that was published in Parallel and was focusing on association with liver enzymes, there was a strong signal on chromosome 22 with this variant 738409, which we subsequently and many, many others have shown is associated also with the histological severity of NAFLD. So, if the previous slide has shown that it's associated with increased liver fat in the general population, this is also associated with the severity of NAFLD with the carriers of the minor allele have worse steatosis, worse inflammation, worse injury and fibrosis. And interestingly, for this example, the inflammation is independent of the steatosis grade. So, for each level of steatosis, you have worse inflammation. This was recently shown in a nice prospective study in Italy that carriers of the minor allele not only do they have worse histology, but actually they have higher risk of hepatocellular carcinoma. This is in a cohort of patients with fatty liver disease, and the risk is greater and more elevated in people who start with baseline advanced fibrosis. Finally, in this nice paper from analyzing enhanced data, it was shown that people who are homozygotes to the minor allele and carry the mutation actually have a higher liver-related mortality. So overall, if we summarize the effects of the PNPLA3 I148M variant, we can see that it's associated with worse outcomes in terms of liver fat, of the histology of NASH, worse fibrosis, higher risk of hepatocellular carcinoma, and higher liver-related mortality. A second strong hit was in TM6SF2, the transmembrane six superfamily member two. This was again picked up in a genome-wide association study from the Dallas Heart Study, and as associated with the amount of liver fat. And you can see in this graph very nicely that carriers of the minor allele limitation from E2K have a much higher liver fat compared to heterozygotes and compared to homozygotes to the major allele. You can also see when you look at the raw numbers that this is a relatively rare variant as previously presented. When looking at the effects on liver histology in this large cohort study of patients with NAFLD or head liver biopsies, you see that carrying the minor allele K increases the risk of higher steatosis grade, higher inflammation, more ballooning or hepatocyte injury, and worse fibrosis. So again, we have a variant that's associated with the severity of NAFLD. Just really fresh off the press and data from the UK Biobank, it was shown that heterozygotes and homozygotes for the minor allele have an increased risk of liver-related mortality in an increasing manner. If, again, suggesting that this is a deleterious variant. And to go back to this table, essentially we have the same pattern of association. The association with hepatocellular carcinoma was shown predominantly in the context of alcoholic liver disease for now. MbO27, membrane-bound O-acetyltransferase domain containing 7 is the third gene I'll mention. This was actually discovered initially in alcohol-associated liver disease, and then confirmed in the context of NAFLD. So this is a study of patients with NAFLD who have available liver biopsies. And you can see that the carriers of the minor allele T have worse steatosis, worse inflammation, and worse fibrosis, although there was no effect on hepatocyte ballooning. This was also shown to be associated with higher risk of hepatocellular carcinoma. Very interestingly, it seems that the risk of hepatocellular carcinoma seen here in the entire population is predominantly driven by an increased risk in the people who at baseline do not have advanced fibrosis, whereas in people with advanced fibrosis, there may be slightly higher risk numerically, but this was far from being significant. And the biology behind this is quite intriguing. So this variant in MBR87 is associated with increased liver fat, with worse histology, with fibrosis, with hepatocellular carcinoma, and to the best of my knowledge has not been demonstrated to be associated with mortality, at least until now. The final individual gene that I'd like to mention is HSD17B13, or hydroxysteroid 17-beta-dehydrogenase type 13. This was initially described in this genome-wide association study from 2011 with a very large, I believe it's about 80,000 people from the general population, and they looked for genetic signals that associate with liver enzymes. PNPLA3 comes up in any study, and there was this signal on chromosome four, which is close to HSD17B13. And when we genotyped this variant in subjects with biopsy-confirmed fatty liver disease from the nurse CRN, we found that the minor allele is associated with more steatosis, but actually less inflammation, less ballooning, less risk of malarie bodies, and a trend for lower fibrosis. So essentially, we have a variant that is associated with more fat, but is protective of injury. Now, as opposed to the variants that I mentioned before, this SNP that was discovered in the GWAS is actually a non-coding SNP. It's located about 12 kilobases downstream of the gene, and it's not affecting the amount of gene expressed, and it's not causing a mutation. However, we did identify another variant that's extremely highly correlated with this one. It's almost a one-to-one correlation. And this variant within the gene region affects the splicing of HSD17B13 and generates splice variants that are enzymatically inactive. And because it's highly correlated, it has the same pattern of association with the minor allele associated with more steatosis, but less injury. So this is another way of looking at it. These are the odds ratios or the adjusted odds ratios for steatosis, which is higher in the minor allele, but improved or protection from risk of all the patterns of NASH. And if we look at diagnosis from the UK Biobank, we see the same thing, that the minor allele is protective from having a diagnosis of cirrhosis, other diseases of the liver, as well as alcoholic liver disease. This was also shown in this beautiful study by Noor Abulhusen from Regeneron, where independently, this was identified in a genome-wide association study as a SNP that's associated with liver enzymes. And the splice variant was shown the same, that there's a trend for a higher presence of that variant in subjects with steatosis compared to control, but it seems to be protective from NASH, from fibrosis, and also from hepatocellular carcinoma. And this has now been replicated in many, many other studies. This nice study in patients with alcohol-associated liver disease has shown that carriers of the minor allele in the splice variant actually have, not only do they have lower risk of developing the disease, they actually have lower risk of developing hepatocellular carcinoma. And in this cohort study from Denmark, in the Copenhagen cohort, it was nicely shown that carriers or homozygotes for the minor allele in the splice variant have a lower liver-related mortality. And even if you correct for cirrhosis and you look at the time from cirrhosis diagnosis, people with the minor allele actually do better. So as opposed to the other variants that I've shown you, here we have a variant that is possibly associated with liver fat. The data is somewhat conflicting on that, but is associated with protection from severe histology, protection from fibrosis, protection from hepatocellular carcinoma, predominantly shown in the context of alcohol, and protection from liver-related mortality. So as Dr. Sukhoian previously mentioned, this is a polygenic disorder, and each of these variants contribute only a little bit to the severity of disease. And since these variants are presumably independent of each other, can we actually combine them to see a genetic risk score? So this has been now published in multiple studies. I'm just giving one example. In this study, what the authors did is they just counted the number of risk alleles in those three genes, in PINPLA3, TM6SF2, and HSD17B13. And you can see in two independent cohorts that the higher the number of risk alleles, the higher the average ALT in the cohort. And the higher the number of risk alleles, the higher the risk of developing cirrhosis, or the risk of developing hepatocellular carcinoma. This will be discussed further by Dr. Valenti later on. We know that fatty liver disease has an interesting racial and ethnic distribution. For example, we know that the prevalence is higher in Hispanics, when people with Mesoamerican ethnicity, and that African-Americans are relatively, or people of African descent are relatively protected from natural decompilation to Europeans. How does that translate to the genetic makeup? Well, I'm showing you the frequency of the lead variant for each of the four genes I mentioned. So the PINPLA, the frequency of the PINPLA3I148M mutant is roughly tracks that of the population prevalence of NAPHLD, whereas for other variants, the story is not that simple. And remember that for HSD17B13, the variant is actually protective. So you would expect the opposite pattern. To the best of my knowledge, a study that actually quantifies the amount of contribution of genetic variants to the ethnic and racial diversity has not been published. Finally, if we have all of those variants that three out of four that I mentioned are bad for us, why are they still so common in the population? Was there anything that's selected for them? So this is data that PINPLA3 protects from gallbladder disease. And this is an intriguing paper or report that literally came out a few weeks ago, which suggests that people who carry the PINPLA3 mutant, although they have higher risk of liver disease, they seem to be relatively protected from severity of COVID-19 disease. So there may actually be beneficial outcomes to carrying the deleterious variants of the genes that we mentioned. And this is something that is under continuing study. So I want to go back to this slide. And interestingly, all the variants that I've shown you, and essentially most of the variants that have been shown to be associated with fatty liver disease for now are here, are either in the very early steps of accumulation of fat or straddling like HSD17B13, the transition from fat to injury. And I think this tells us something about how this pathogenic slide probably captures the true pathogenesis of disease, as well as where we should maybe should be focusing our therapeutic efforts. So I'll conclude by summarizing that I've shown you that there are common genetic variants that influence the risk of developing fatty liver disease, its severity and meaningful outcomes, that most of the genes that are found to be associated with NAPFLD are involved in lipid metabolism, and that if we identify those associations, we may be able to identify pathways that we can then target therapeutically. And with that, I will leave the stage to Dr. Valenti who will talk about the therapeutic options and the clinical implications. So the main unmet clinical need in the field is probably the identification of effective treatments and targeting genetic risk variants after the opportunity to correct disease pathways in homogeneous subsets of patients. The main unmet clinical need in the field is probably the identification of effective treatments and targeting genetic risk variants after this opportunity. The obvious candidate is the PMPF3-148 metamine variants. As we have heard, it has been identified in genomic studies as the major common genetic determinant of the fatty fat accumulation in our population. And not only are the fatty fat, but all histological hormones of NASH at genome-wide level as highlighted in this recent study by the NAPFLD, studied by Quentin Anstey and co-workers which has been published in the Journal of Pathology. So this is a rare example of a common genetic variant associated with a moderate effect on a trait and the main common genetic determinants of this disease. So as we have just heard, the global prevalence of carriers of this variant mirrors that of the NAFLD desire in Hispanics and decreases progressively in Asians, Europeans, Africans. And the variants accounted next also for a larger fraction of hepatic fat variability in the multi-ethnic Dallas study where the variant was originally identified. Even in Europeans, we recently show and the population base UK Biobank cohort encompass the evaluation of more than 350,000 individuals from the UK that this illness in 148 metering variant accounted for as much as 16% of cirrhosis variability and almost a quarter of variability of the success ability to develop hepatocellular carcinoma. Particularly striking is the relationship between homozygosity for the variant and the hepatocellular carcinoma. With an almost ninefold enrichment in cases as compared to healthy European individuals without fatty liver disease. You can see that the prevalence of homozygosity is about 5% in the general population without liver diseases compared to about 35% in those who develop hepatocellular carcinoma due to NFLD or alcoholic liver disease. So this is what would suggest that correcting the molecular pathways underlying the liver phenotype associated with this mutation can reduce the burden of liver events as also confirmed by the studies shown before suggesting that this variant has also a consistent effect on mortality in patients with fatty liver disease or dysmetabolism. Concerning the underlying mechanism responsible for liver disease development, we know that the main trigger for the phenotypic expression of the variant is represented by excess adiposity probably for the induction of insulin resistance at least in non-heavy drinkers. We know that the wild type PMPR3 protein is induced by insulin at least in mice and it is expressed at the surface of lipid droplets including in hepatocytes where it is involving the remodeling of triglycerides and phospholipids by particularly acting on unsaturated and polyunsaturated fatty acids. And then it's degraded. Through a series of elegant experimental studies, Ellen Hopps' group has shown that the 148th methionine variant will lack this enzymatic activity in mice accumulating lipid droplets acquiring the ability to sequester ABHD5 which is an essential co-factor for ATGL. The main triglyceride lipase in hepatocytes into lipid droplets enlargement due to the entrapment of lipids of triglyceride within the lipid droplets and probably triggering liver disease by inducing lipotoxicity. We should also remember though that the loss of function of enzymatic activity impairs also the remodeling of phospholipids with unsaturated fatty acids which may affect the intracellular membrane composition and affect the liver disease by this mechanism. But if protein accumulation is essential for the detrimental impact of the mutation, then we can hypothesize that liver-targeted PMPLA-free silencing may offer a therapeutic strategy as shown in the panel on the right to treat patients homozygous for this powerful response for the disease. Importantly, this therapeutic approach seems to work in mice. In the study shown here, the author showed that the PMPLA-free silencing in mice by antisense oligonucleotides or ASO was able to reduce hepatic fat accumulation in mice fed a steatogenic diet as shown in the bottom. Here, the fat is shown in red by the aurostiming. And the effect was larger in mice and generate to express the mutant PMPLA-free on the left as compared to mice expressing wild-type PMPLA-free. Remarkably, PMPLA-free silencing was also able to prevent fibrogenesis as shown on the top and fibrosis accumulation as shown by immunochemistry in the middle panel. And again, the impact was larger in mice over-expressing the mutant PNPLA3. So this offered a first proof of principle that a selection of genetic risk variants associated with disease susceptibility followed by the characterization of the underlying mechanism can link to the development of a new precision medicine approaches to modulate these genetic disease pathways, which will still need to be validated in the clinics, but may be suitable for clinical trials already. Furthermore, as shown on the right, PNPLA3 expression can be regulated also by other ways, for example, by a modulation of signaling pathways acting on its transcription by regulation of autophagy at protein levels or indirectly by silencing or inhibiting, for example, HSD17V13, because a genetic epidemiological study that has shown that the impact of this mutation on the protection of liver disease is particularly evident in carriers of the genetic variant of PNPLA3. And the first phase one clinical study evaluating a direct HSD17V13 inhibitor has already been present and showing that this compound is safe in the short term. And it was associated with a reduction in non-invasive biomarkers of liver damage. But a strongly interconnected aspect is the identification of patients at higher risk of liver progression, which we need to treat to prevent this liver-related complications, which may be based on the rapidly increased number of genetic risk variants for fatty liver disease, which is increasing by the day. And these are common genetic risk variants for both non-alcoholic or metabolic dysfunction associated fatty liver disease. And also, as we have heard before, alcoholic fatty liver disease. The discovery that the impact of the main risk variants on severe fibrosis and hepatocellular carcinoma is proportional to their impact on hepatic fat accumulation has allowed us to use these genetic risk scores for an FLD to predict liver-related complications of this condition. For example, we recently show in a large clinical cohort of European individuals and a population-based UK biobank cohort that a score summarizing the impact, the weighted impact of the main common genetic risk variants for this condition are the most robustly validated as highlighted before in PMPLA-3, M6-SF2, MBUS-7, glucokinase regulator, NHSD-17B13 was able to predict not only the development on FLD in at-risk individuals with this metabolism, but also the progression to cirrhosis and independently of clinical risk factors and also of cirrhosis, the further progression to hepatocellular carcinoma. Importantly, this was also able to predict robustly the evolution to hepatocellular carcinoma, the development of hepatocellular carcinoma at least in patients without advanced liver fibrosis. The discriminative ability of this polygenic risk scores of hepatocellular carcinoma in particular was superior to the predicted ability of single genetic risk variant, including PMPLA-3 in blue or a combination of PMPLA-3 and M6-SF2 in yellow on the left. And it was also superior when combined with classical risk factor to clinical risk factor alone in the ability to discriminate the risk of hepatocellular carcinoma in a clinical cohort, independently of classical risk factor. And it should be noted that among the clinical risk factor for this condition, they were considered age, sex, BMI, type 2 diabetes, but also the presence of advanced liver fibrosis, leading to a correct reclassification of about 20% of the patients who were not correctly predicted to develop hepatocellular carcinoma based on clinical risk factors alone. So this will require further prospective validation, but we expect that the progressive addition of new genetic risk variants for an FOD as the two recently identified risk variant in JIPAM and APOE in the UK Biobank, and then validated in several clinical cohorts will further improve this course, leading to an improvability as shown on the right to discriminate the risk of fatty liver disease, but also of liver-related events related to fatty liver disease in the population. So how to implement these findings in clinical practice? Currently, in at-risk individuals, the first screening step with evaluation of fibrosis score, such as the FIF-4 is advised, followed by evaluation of liver stiffness measurement and the use of other patented score of liver fibrosis to identify individuals at high risk of liver-related complication. But this still leaves a gray area and will require the characterization of very large number of individuals with abnormal fibrosis score, more in details. Our group has recently shown that the previously reported polygenic risk score could further stratify the risk of liver events, both in patients with high and intermediate fibrosis score, such as the FIF-4 in the middle upper part of these slides, and the NFLD fibrosis score, leading to a refinement in the prognostic classification of this patient. And this data were obtained, again, in the UK Biobank Core. And the discriminative ability on the risk of liver events was also independently confirmed by other groups in Finnish population, as shown on the left. Again, the impact of the high PRS on the risk of liver-related events, with a larger effect in individuals with NFLD as compared to those without NFLD. Suggesting, therefore, that evaluation of PRS may be useful in individuals with fat liver of risk factor for this metabolism with abnormal FIF-4 values. So in these patients having high PRS or being a mosaicus for the PMPLA-3 variant will highlight an increased risk of liver-related events as compared to cardiovascular events, which will have implication for clinical management and surveillance for complication liver disease, but may also highlight the utility of treatment with specific approaches, such as drugs targeting the PMPLA-3 or pathway or HSD-17 and B13 in the future. So finally, would this be sufficient to allow a precision medicine approach? Probably not yet, as accumulating evidence suggests that besides the rare combination of common variants leading to a high polygenic risk scores, rare genetic variants with a large effect may also be responsible for severe NFLD, also in adult patients, especially in those without strong environmental triggers for liver disease, such as obesity with type 2 diabetes. For example, we show a significant enrichment of mutation in apolipoprotein B associated with a phenotype consistent with hypo-beta-lipoproteinemia in Italian patients who develop parathyroid carcinoma due to NFLD, both in the presence and the absence of obesity. So that whole exome sequencing or genomic evaluation in patients with severe phenotype related to NFLD, especially if unexplained by Aquarius factors, such as patients with severe NFLD without overweight, has been proposed to identify the presence of mutation causing not only, as shown on the right of the slides, hypo-beta-lipoproteinemia, but other lipid metabolism disorders as well as the Canary-Dolphman syndrome, love deficiency, lipodystrophies, mitochondrial disorders, or other genetic liver disorders as well, which will have clinical implication not only for family counseling of the affected problems, but also because some of these genetic mutation are amenable treatment. For example, the supplementation with liposoluble vitamins for hypo-beta-lipoproteinemia or leptin replacement for some form of lipodystrophies, replacement therapies for lysosomal disorders, and so on. So with this, I thank you for your attention. I now hand it over to Professor Diehl for the Q&A. Thank you. Thank you very much. Panelists, this has been a great discussion. We continue to try to overcome technical obstacles here. I'm not sure why you can't see me, and I can see Luca, and I can see Sylvia, and Yaron. So they're the important ones. We did not get any questions in the question and answer period. So I guess this gives us an opportunity to maybe raise points that you think are the most worthy for further contemplation. Let me throw it to you, Sylvia. You gave a nice overview of the field. Where do you think this is going? What do you think should be focused on from a clinical practice point of view? Oh, thank you, Anna-May. Can you hear me? Yes. Okay, good. Certainly, we need to know more. As I introduced the concept of misinheritability, I do believe that the field is still evolving. We don't know all variants involving the susceptibility or the protection against the disease. So in order to know where are we going, we need not only to know which are the complete spectrum of variants modifying the disease, but also we need to know more about the real involvement of each of these variants we all mentioned in the disease and how the protein exactly works in the frame of the disease because as you review the literature, you can find more papers coming and coming about PMP-L3. But as Yaron and also Luca mentioned, we don't know exactly how the other genes or proteins exactly work because we don't know nothing about the ligands of the proteins. So we don't know how the proteins interact with other proteins and also which are the specific ligands. We cannot predict future targets. So I think the field is evolving. We know many things, but we still need to know more. I don't know how is your view, Luca or Yaron, but to me, we need to learn more about this. Yeah, I think your response is quite timely because I do see we just got a question from one of the audience asking whether the polymorphism was a gain or loss of function. And maybe one of our experts can address that. What does this thing do? And I myself ask if it's affecting fat and we know that there's a poor correlation between the severity of steatosis and the outcome of the disease. How do we get from fat to cancer? I'm just asking. I think it gets back to loss or gain of function and what function. So I think that's an excellent question. And also I'm not sure that the answer is only those two options. For example, there is at least in vitro data for the I148M may actually, may affect the ubiquitination of the protein. And then it actually, it's through its interaction with other proteins. So it may not be a loss of function or gain of function of the enzymatic activity of PNPLA3 itself, but may actually be a modifier of other proteins, for example, through CGI58 and ATGL. I also don't know that associations, to answer your question, I don't know that associations with liver fat directly drive injury. It may be that the associations are with the processes that are upstream of liver fat. So something that is associated with the drivers of liver fat may be associated with the drivers of injury. But, and I completely agree with Dr. Sukhoian that we need to understand the actual biological mechanism of the proteins that we're dealing with in order to understand, I think, more about the disease. Yeah, I also completely agree. I mean, we made the huge steps as compared to 10, 15 years ago, but we still to know much better, to understand much better the mechanism in order to develop more effective therapies in my view. These are good to design the first course because we don't know to know exactly the mechanism to use them as biomarkers, but for therapeutic manipulation, we need to understand better the mechanism. Let me, some people are also picking up on the complexity of this. I was struck by the protective effect of the polymorphism on biliary tract disease. And one of our audiences asked, it seems to be a protective marker for COVID. So any thoughts about that? I think we are all mute in the mind, not in the computer, because it's something very, very new. But let me say my thought about this. As I said in the beginning, we need to learn more. And this is specifically regarding when you design specific therapy against any of the targets at any of the body, because the off-target effects can be one of these ones. For example, you see the PMPL-3 as an excellent, a good candidate, perhaps the best one for the fatty liver, but now COVID-19 is in the air, literally, but it is now fighting as a new player. I don't have an answer, but probably we can start looking for the effect of this PMPL-3 protein and the hormonal effects on fat. It is my guessing. I don't know. I don't know exactly the answer, but as you can see, the only transcriptional regulation network of PMPL-3 is really, really huge. And the epigenetic modifiers related with PMPL-3 are also very important. So when I say these things, the diseases in which the variant is involved in protection or susceptibility, I feel we don't know anything because we are still trying to understand this, but I don't know how are the thought of Luca or Yaron about this. Well, I agree. I mean, I can confirm that we also see protection against severe COVID in our court in Milan in case of the variants. So it's consistent with the published data. I don't know about the Galston disease or cholesterol, phospholipid metabolism. I mean, we should study it more, but I really agree on the fact that any therapy will be associated with targets of fat and side effects. So we should study it better, understand it better before proposing it for therapy. Well, thanks. I think we've done a great job. I think it was a fantastic seminar. I learned a lot. I hope our audience did. We do have one question that maybe someone can type an answer to because we are literally out of time. Someone questioning whether MBOT7 is really an Apple risk factor. So if you guys can respond to that via your chat box, I think we are out of time, but I'd like to thank all of the panelists. Just you did a great job. It was fantastic. And I hope the audience enjoyed it. Looking forward to our next webinar from the Naflatzschig. Bye. Bye. Thank you.

genetic variants

non-alcoholic fatty liver disease

NAFLD

PNPLA3

TM6SF2

MBOAT7

HSD17B13

fibrosis

targeted therapies