Good morning, everybody. Good morning. Why don't we get started? My name is Don Rockey. And on behalf of my co-moderator, Betsy Verna, it's a pleasure to start this session. I have to say, in getting started, it is great to see so many people here. So thank you all for getting up. This is a joint program between the Portal Hypertension SIG and the ACLF SIG. And so I'd also like to acknowledge the ACLF troops. So they're up here in the front, Andres Cardenas and Dean Carvelos. Where's Dean? Dean's somewhere around. We spent a fair amount of time trying to organize this. But the goal is to provide value to you all and get you on a trajectory from early portal hypertension to sort of the end stage in the hepatorenal syndrome. Just a couple of comments about the program today. So we're going to go from 830 to about 945. We're not going to have questions during the talks, but we're going to have a panel at 930, 15 minutes. Then we're going to break for a little bit. Then we're going to have our business meeting, the portal hypertension business meeting. And then at 11, we're going to come back for the ACLF component. And that's going to go until 1230. And then they're going to have their business meeting. So we'll try to reiterate that so everyone's not confused. So without further ado, I'm going to turn it over to Dr. Verna. OK. And I'm thrilled to kick this program off with Dr. Andres Duarte-Rojo, who's going to be talking to us about assessment of clinically significant portal hypertension, or CSPH. Where did I? Oh, I have to go here, right? Yeah. That's Tom. That one. Mm-hmm. And then from beginning? Mm-hmm. Sure. Go ahead. Yeah, you're good to go. OK. OK. Well, thank you, everybody, for being here. And thank you for the organizer for inviting me to give this presentation. So we're going to go over assessment of clinical significant portal hypertension through noninvasive liver disease assessment. So I'm a transplant hepatologist at Northwestern. These are my disclosures. And so these are the objectives. We're going to first just do an overview of CSPH and what are its surrogates, HVPG, hepatic venous pressure gradient. We're going to talk a little bit about NILDA, how it can surrogate some of the HVPG measures, overview, liver, spleen stiffness. Then we're going to go into the direct measurement, just a little bit, a little taste of that. And then we're going to wrap up. So let's just go over a little bit of the overview. So we shouldn't forget that as we take care of these patients, there is a whole spectrum. And that in the end, let me see if I can get my cursor there. OK, so it's to the left. So we shouldn't forget that we really start with this advanced chronic liver disease. Because some of this CSPH can actually happen in this stage or even in that stage. So we're truly looking into these stages to try to identify these early stages of portal hypertension. So in the end, when we do the HVPG, which is just very similar to the changes or the gradients that we will do in a swine GANS, we're truly trying to identify these two thresholds, above 10, clinical significant portal hypertension, or above 12, which is the patients that will be at risk for high-risk varices or risk for bleeding. Going into 16 and 20 really hasn't been that fruitful in the studies, so we're not going to truly touch onto that. And just as a little reminder of the anatomy of portal hypertension, or its classification, sorry, how you have intrahepatic, posthepatic, prehepatic, and the intrahepatic, you can have the sinusoidal, post, and pre as well. Not all of them we're going to be able to identify through non-invasive liver disease assessment, or NILDA, which is how I'm going to call it from now on. So there are two main NILDA techniques. Basically, they look into biological properties or physical properties or tissue that will make you aware of how much stiffness, how much fibrosis, or how much steatosis, or whatever you decided to assess, it's in there. So you can either do it from a blood sample, where you look for those biologic parameters, put them together to estimate whatever you want to estimate, or you can do it through imaging-based. In the end, and for this talk, we're going to really only refer to those studies that look at the comparison with the HBPG, as I described before. And we're going to touch a little bit into the outcomes of portal hypertension, which perhaps is a new paradigm that we need to be exploring a bit more. So the standard of routine imaging is not good to identify portal hypertension, or it's not good at the earliest stages. Because once you see a dilated portal vein, once you see all these portal systemic shunts, once you see a 15-centimeter spleen, the train has departed. It's already late. We really want to identify patients before they get to this stage. And that's where, perhaps, imaging-based NILDA is what is going to be more useful for this group. Whether you do it through mechanical vibration or acoustic vibration through a regular ultrasound, a regular B mode, or also through elastography, in the end, basically what we're doing is like if we were just tapping on a watermelon and just hearing the sound that it comes from it and determine how stiff the tissue is. Spleen stiffness has the same principles. And again, you can measure it either through transient elastography, TE, through SWE, shear wave elastography, or through magnetic resonance elastography. So a little bit on blood tests. And one of the things that actually is coming in the guidelines that have been submitted to hepatology, the guidelines on NILDA and portal hypertension that were actually directed by Don Rockey. He's a senior author on that paper. Is that platelets are really not a good surrogate. And that if we're basing our HPPG measurements or basing clinical significant portal hypertension or platelets, we're not doing a good job to our patients. And then we have all of the other markers. And they're not really that sensitive or specific with a range of 50% to 70%. So again, we have to do something better in terms of blood-based NILDA. I like this study that I want to show to you, where they actually found a way of boost up the FIB4. I'm sure you're all aware with the FIB4. It includes AST, platelets, ALT, and H. And by adding albumin and just getting into this formula, they actually were able to have a good area under the curve or a good accuracy to identify patients who have portal hypertension. You can actually go into their website, and it will tell you what the probability is for having CSPH. They actually came up with these two code of values. Your FIB4 is above 6, and your albumin is below 3.6. Your patient has a 75 probability of having CSPH. So I think this is something that we can easily use in the clinic, easily incorporate, and give us enough sensitivity to get started with our patients. The ELF, I believe it's a promising marker. Not all of the studies have been positive. But this is a good study where I just want to show you that its accuracy was actually not off from a transient elastography. So I think we need to hear more about this biomarker. Even though it did not make it to the guidelines, I think as more information gets accumulated, this might be something that we will be using in the future. So let's go into the imaging base, and I'm going to start with the liver stiffness. So the rule of five is what came in Babenos 7. And in order to get there, which I'm basically just putting here what the rule of five represents, in order to get there, it's important to understand that we have the rules from Babenos 6, how to identify compensated advanced chronic liver disease, or CSPH, some rule out thresholds of 10 kilopascals, ruling thresholds of 15 kilopascals. Then this study showed us that perhaps 8 and 12, respectively, are better cut off values if we want to identify CACLD. We came with the Babenos 6 favorable status based on a study, which I'm going to show in a minute, where if the LSM is less than 20 and the platelets are more than 150,000, this is a patient that perhaps you do not need to do endoscopy, and you can just obviate that and continue doing the rest of your care. That actually came from the anticipate model, and that's what fed into the Babenos 6 criteria, which has been evolving into more important results and now into the rule of five. Within that, I just want to show you the areas under the curve for three different methods. And even though the LSPS was the best, the liver stiffness had a very good area under the curve. And when they combined it to platelets, that's exactly when they got just the best results. With the LSPS, the data have been less consistent, and this is why also, even though it's described in the upcoming guidelines, it did not make it to the recommendation of something that is just going to be practical and accurate enough to use in the clinic. So going into the rule of five, and this is an important study that also evolved from the anticipate and now created the anticipate NASH model. And what I want you to point attention here is that using this rule of five from 10 to 25 or 15 to 25, if patients have less than 15 kilopascals of liver stiffness, you have a pretty high negative predictive value that you actually made it almost perfect 97% by adding platelets. The positive predictive value of 25 kilopascals is also good, except for patients that have a BMI more than 30. But if you incorporate the BMI into this formula, you can also get to very similar positive predictive values. So this is a large study where over 1,000 patients were included to prove the foundations of this rule of five. And what I want you to pay attention is that they use this exclusion of CSPH when patients have less than 15. A ruling when patients have more than 25 kilopascals. And they also use this low or high intermediate groups where patients have 15 to 20 with less than 110 platelets or 20 to 25 with less than 150 platelets. And you can see that the ruling out of less than 15 and 150 was actually pretty good because none of the patients decompensated. Just 1% had death. And whereas the group with more than 25, 15% decompensated, there was death in 6%. The intermediate was also did what it was supposed to do, but only a 3% of the compensation rate. And when you actually put this into this Kaplan-Meyers, you can see how the high and low probability for the intermediate group, they kind of go together, except for the prediction of death where the high probability kind of moves more into the patients that had confirmed CSPH. You can see here the hazards ratio for each of those outcomes. And what was also interesting is that when they assessed the rule of five, the only one that came as a statistically significant was more than 25. So I think this is a number that we need to remember to just ruling in clinical significant portal hypertension. And this was also significant in their multivariable analysis with other parameters. Now, we have a problem with NASH as the anticipate study showed or the anticipate NASH study showed, which is that we can get the CSPH in patients when they have F3. We can have the compensation events in patients without CSPH, 8% of them. And there is underestimation of the wedge hepatic vein pressure as shown in this study, where they compare a direct puncture into the portal vein versus the wedge. And you can see how in patients without NASH, this is pretty tight in this plant outment. And here, the fact that there is dispersion just tells you how there is inaccuracy. And in the end, what happens is that the wedge is underestimated in 33% of patients with NASH. And so I mentioned before that perhaps we need to start looking more into outcomes. And I wanted to bring this study from the cell insertive CYMP2SUMF data that actually shows that if your stiffness is over 30%, you have a higher risk of decompensation. You have a higher risk of mortality. So perhaps what we're going to be looking is that in patients with NASH, we're going to need to consider this study. We need some confirmatory studies. But you also need to consider that if the patient has active CLD, you have to be careful with the code of value. Sometimes they might be just going up because there is a modulator or a modifying co-factor. And we're going to talk about that in a few minutes. A little thing on SWE. And this is a study that compared TE with SWE. And as you can see, there is almost exactly the same results, very similar accuracy. And you can see that there is also almost the same code of values that you can use to identify ACLD either with SWE or with TE. This is another study where they compare the HVPG versus 2D SWE. And you can see that the R's are pretty good. They're as good as they have been with transient elastography. So basically, and it's also showing less variation across etiologies than transient elastography. So most likely, we can use exactly the same code of values that we've been using for TE with 2D SWE or with RFE in general. As we move into MRE, I just want to show briefly a couple of studies. And you can see these areas under the curve. They're really, really good, over 88% to 90% for liver stiffness and for spleen stiffness as this code of value for liver stiffness of 5.05. And the same was later on proving in other studies of the MEVFIP, where patients who started to develop decompensations were those that precisely had over 5 kilopascals on the MRE. So that's another code of value that we need to consider. I'm going to talk a little bit about spleen stiffness. And the issue with spleen stiffness is that we have too many code of values. The studies are smaller than the studies that we have for liver stiffness. And there is rapid changes in technology. This study perhaps summarizes most of what I want to present to you, that if you do a single code of value using 40 of spleen stiffness versus two code of values using 21 to rule in and 50 to rule out, sorry, and 50 to rule in, this 40 code of value has a smaller gray zone, has a higher rule in zone. And of course, it was better compared to the other strategy of using dual code of values of 21 and 50. And of course, it's also a bit better than the LSM. So likely, this Babeno 7 that is being proposed along with the spleen stiffness is what we should be using and especially using this code of value of 40. This is a more recent study where they analyzed the new probe. And you can also see here that the spleen stiffness area under the curve is way higher than the liver stiffness. And also, with this code of value, you can prevent some more endoscopies, up to 78% of endoscopies versus 53 when you use the code of value of 40. So there's more to be said. This is what we do at Northwestern. Whenever a patient has above 12, we automatically do the spleen stiffness. But whenever we suspect that there is a presynocytal cause, then we ask our technicians to do an extra SSM. So this is just a patient with alcohol-associated liver disease. You can see the liver stiffness versus the spleen stiffness. And this is a patient in whom we have to request that extra spleen stiffness and show the portal hypertension despite the patient having no cirrhosis. Very briefly, on direct portal pressure, instead of having to get everything in a Swan-Gans model, we can just directly through endoscopic ultrasound puncture those veins of interest and get a direct portal pressure measurement. These are the studies that have been published, very small samples. This is the correlation, very small sample. However, it does do what it's supposed to be doing, which is predicting the presence of viruses. So as we make sense of all results, I think we can use these values of 25 to rule in the values of less than 15 on liver stiffness to rule out. Use the spleen stiffness cutoff value of 40 also to detect clinical significant portal hypertension. And we still need a little bit more data on how we're going to longitudinally assess changes as we get repeat measurements. This is what you're going to see on the guidelines, where it's especially making emphasis to these cutoff values that I'm showing to you. You have to use your clinical judgment for the values between 15 and 25. And there are multiple modifiers. One of them, you can see them here. One of them is going to be congestive heart failure. This is just a patient. Briefly, I want to show you how there is a dilated hepatic vein. This patient with fallacy tetralogy, after she got valve replacement, you can see the improvement in the cardiac function. And you can see also the drop in the stiffness. So NILDAP can help identify clinical significant portal hypertension. I think FIP4 and ELF are promising. We should incorporate them, especially this FIP4+. And the LSM and SSM Babenow algorithms and the Rule of Five are clinically our best guide so we can enhance our clinical care. So beware of the modifiers, and where available, the endohepatology is a valuable resource, but we need to learn to use it better. Thank you very much. Thank you. Okay, great. I'd like to welcome our next speaker, Marina Serper from Penn, who's gonna give us a lecture on is it possible to prevent decompensation in patients with established fibrosis and portal hypertension. Sorry. Okay, thanks a lot, everybody. So I could have just made one slide and said the answer is yes, but it's a little bit more complicated than that. So let's talk about this. Here are my disclosures. So we'll talk about portal hypertension as a key therapeutic target to prevent hepatic decompensation and the paradigm shifts that we have recently accepted. I will talk about pharmacologic therapy with mechanistic and clinical evidence, and there's a lot more mechanistic than clinical. I will focus mostly on non-selective beta-blockers, and in particular, carvatalol. Then I'll talk a little bit about anticoagulants, anti-factor Xa, anoxaparin, direct oral anticoagulants, antiplatelets, aspirin, and a little bit about combination therapy to prevent decompensation, and we'll outline some areas of unmet needs and future directions. So we are now interested in this concept of advanced chronic liver disease, but first, of course, we accept the paradigm of compensated and decompensated cirrhosis. Compensated cirrhosis being asymptomatic, having a long-term survival, and decompensated cirrhosis being marked by complications, variceal hemorrhage, ascites, et cetera, with a much shorter survival. And you heard from Dr. Duarte-Rojo that we're interested in CSPH, for which the gold standard of measurement is transjugular venous pressure measurement, and an HPPG of greater than 10. And we know that CSPH increases our decompensation risk. In our new paradigm that we've accepted now is we have the concept of advanced chronic liver disease where we need to prove that a patient has cirrhosis either by imaging or biopsy, and we accept that cirrhosis and advanced fibrosis are on a continuum, and patients with advanced fibrosis could have CSPH and decompensate, and we need to follow these patients. We could also assess CSPH non-invasively, as you heard from the last speaker. And we really wanna reduce portal pressure as a therapeutic goal. And so an important concept for me and for others is that we are actually trying to prevent largely ascites. Ascites is by far the most common decompensating event. We had previously accepted the fact that we can use non-selective beta blockers as primary prevention for variceal hemorrhage for large varices, but now we're really thinking about preventing ascites. And it's been shown that ascites is the most common decompensating event in both alcohol viral populations and those with MASLD in prospective cohort studies. And so I'm now going to repeat how we identify patients with compensated cirrhosis and CSPH, but obviously we have our gold standard, HVPG. We can also identify them by portosystemic collaterals or varices, and also NILDA and platelet count, liver stiffness, et cetera. So once we have these patients in front of us, let's talk a little bit about therapies and what's ready for prime times. And this is not necessarily to prevent a first decompensation, but just what we have in our armamentarium. I think it goes without saying, we all understand that removal of the etiologic cause is extremely important. So treating underlying viral causes, losing weight, removing alcohol as an agent is going to be important. And I won't focus on that so much. I won't focus on any sort of mechanical therapy. I will talk a lot about targeted pharmacotherapy. And we'll talk a little bit about emerging therapies. We have statins, anticoagulants, and antiplatelet agents. These are agents that a lot of our patients take for other reasons. And so there's an opportunity there to study this in observational studies and also in clinical trials. And so these are the agents that I will focus on today. I won't focus on anti-inflammatory agents or antifibrotics. So let's first talk about non-selective beta blockers. These have been around for many, many years. And so the proposed targets here in terms of portal hypertension pathophysiology are with beta one blockade, we are lowering the cardiac output. With beta two blockade, we are constricting the spliconic vasculature. And then with alpha one blockade, we are lowering the intrahepatic vascular tone. And this is where carvatalol has additional mechanisms beyond propranolol and natalol as an alpha one adrenergic blocker. And so we of course need to discuss the landmark PRADESHI trial, which enrolled about 200 patients in a prospective manner, evaluating propranolol or carvatalol on hepatic decompensation and death. And so patients all had a CSPH and HPPG of greater than 10. Those who were responders to propranolol, stano-propranolol, those who were non-responders were given natalol and followed longitudinally. And so what they found in this landmark study was a death or decompensation was 27% in the placebo arm, 16% in the NSPB arm. There was noted a larger reduction with carvatalol and more than propranolol in HPPG. And the biggest effect in terms of decompensation, again, was ascites with a hazard ratio of 0.51. And please notice that the effect was really noted after two years. And then in subgroup analyses, effects were consistent largely across subgroups, though patients with small varices did seem to benefit more, as well as patients with alcohol-related liver disease. And because we don't have a lot of prospective data, we actually simulated this trial using a veteran cirrhosis cohort. And so we did an active comparator new user design of patients with cirrhosis who had very similar criteria as Pradeshi, no prior decompensations. And we looked at new users who were started on selective beta blockers versus carvatalol, largely for cardiac indications and hypertension. And we actually found very similar hazard ratios to Pradeshi. So Pradeshi was 0.51, we found 0.56. Now our subgroup analyses showed some differences. We noted, for example, that patients who were older did not have a benefit compared to younger patients. And we also saw more benefit in child pew six. But again, this was very interesting. We did a lot of work using inverse probability treatment weighting to really balance these groups on the two different types of beta blockers and found very similar results. There has also now been a four trial individual patient meta-analysis that has evaluated carvatalol on decompensation risk and mortality. One of the four studies is Pradeshi. Two other studies actually were comparing beta blockers versus band ligation for preventing ovariaceal bleeding. We do find the very similar signal of carvatalol reducing death and decompensation with a hazard ratio of 0.5, 0.42 respectively. And then finally, just the final piece of data, a very elegant study, a Bayesian reanalysis done by Dr. Rowan Villanueva, where they actually try to use Pradeshi data and use different assumptions to see what is the probability that beta blockers have a benefit in decompensation and to try to quantify that benefit using micro-simulation. And so what they found, that there was a greater than 90% probability of beta blocker benefit, translating at 10 years to survival benefit of about six to 20 months. So I think that's an important point. And then their final hazard ratio for decompensation was in the range of 0.5 to 0.7 in terms of lowering decompensation. And so here's a summary of the current guidance. This is taken from the recently published ASLD portal hypertension guidance, juxtaposed with Bovino 7. So treatment with non-selective beta blockers should be considered for the prevention of decompensation. In CSPH with Bovino, both Bovino and ASLD agree that carvatalol is the preferred treatment of pH in patients with cirrhosis. So now let's move into an area that is emerging. It's very, very interesting from a pathophysiologic standpoint, but still where the human data is emerging. So there's a lot of really interesting pathophysiology of microthrombi formation, platelet aggregation and clotting in this hepatic sinusoids. And there are multiple investigators now looking at the role of endothelium and how that leads to hepatocyte destruction, parenchymal extinction, further fibrosis. And so this is an exciting area. In terms of human data, unfortunately we don't have that much prospective and high quality human data for anticoagulation. So I still have to bring to you the Villa study from 2012 with anoxaparin. So basically patients with child B7 to C10 cirrhosis, about 70 of them were randomized to anoxaparin 4,000 units per week versus placebo. These were all patients free of portal vein thrombosis. They were treated for about a year. And then what you will see is that patients treated with anoxaparin at 48, 72 and 96 weeks were completely free from portal vein thrombosis. Though you will notice that the sample size is waning over time. And this is compared to about one third incidents of portal vein thrombosis in those in the placebo arm. What was also interesting that within the first year, hepatic decompensation was 12% with anoxaparin versus 59 with placebo. And bacteremia or SBP was 9% with anoxaparin versus 33% with placebo. We again used VA data to try to evaluate the impact of anticoagulation, given this data using a VA cohort. And what we did was we took compensated patients with cirrhosis who had atrial fibrillation and were newly treated for atrial fibrillation. And we balanced them on a host of factors, MELD score, child pew, demographics, platelets counts, CHAD-VASc score and use of other antiplatelet agents. So we really tried, we generated two cohorts, DOACs versus placebo, warfarin versus placebo. So what you will see here in the green boxes is that the green boxes are statistically significant results. And these are incidents per 100 person years. All cause mortality was lower with DOACs compared to no anticoagulation. Same with warfarin, whereas you will see we had a higher sample size and more power. We did not find this with DOACs and I don't know if this is a sample size issue, but with warfarin, we have significantly lower hepatic decompensation and death after hepatic decompensation, which was our surrogate for liver related mortality. Splenic thrombosis was a little lower with warfarin, but I don't think we had good data capture because it's difficult using administrative data. There were no differences in bleeding. And then we did additional models looking at time dependent confounding. However, we did find a benefit with both DOACs and warfarin or on hepatic decompensation. These are retrospective data. We can't get rid of all confounding, but interesting signal here. And next I'm gonna switch over and talk about platelets. I think platelets are really, or antiplatelet agents are really interesting targets. There's a lot of interesting in vitro data for antiplatelets, but let's go through the human data. So there's this landmark study by Dr. Simon who actually was looking at aspirin as chemoprophylaxis for HCC in a large Swedish cohort with viral hepatitis. And so what they found was that over 10 years, the absolute risk difference of patients who were treated with baby aspirin versus not was 4% lower for HCC, 7% lower for liver related death. And here you have liver 10 year related mortality on the right. They also showed a dose dependent effect. So greater than one year of aspirin is when they saw a benefit and they saw more benefit with two, three and four years. And if you look at supplemental table 12, which I did, they also evaluated hepatic decompensation. And so this was 19.8% with placebo at 10 years, 11.4% with baby aspirin with a hazard ratio of 0.71. There are also some studies more recently done in the UK Biobank, in the Penn Medicine Biobank, trying to carefully match patients who are treated with aspirin versus placebo that showed a lower risk of new liver disease, lower risk of fibrosis, lower risk of mast cell. And again, Dr. Simon looked at the MGH Naflde cohort and she looked at APRI scores, FIB4, Naflde fibrosis scores. These were lower with regular aspirin users over time. So this is a table that's showing you a lot of very interesting mechanistic preclinical data. There are some human studies, but this is kind of the landscape of what's going on. There's a lot of animal data that's pretty compelling and pretty interesting in fibrosis reversal, but really no human prospective data. I said I wasn't gonna talk about statins, but I wanna just mention one combo trial of simvastatin plus carvedilol versus carvedilol. This was a trial of about 200 patients where about 60% of them were retained. So the patients with an HVPG of greater than 12 were randomized and treated for three months with carvedilol versus carvedilol simvastatin. In the per protocol analysis, there was an HVPG drop of about three millimeters of mercury in each group. However, there was no difference in the hemodynamic response, which was defined as a greater than 20% drop in HVPG or HVPG less than 12. I am not sure that this study was powered to find a benefit because there was about 60 patients per arm after three months. There was no difference in worsening of ascites at three months, but again, this is a short-term study. And I think more studies that are a larger multicenter need to be done. So this is summarizing the current guidance here that I'm not going to read to you, but we're definitely there in terms of recommending non-selective beta blockers. In terms of anti-coagulation, DOACs, anti-platelets, anoxaparin, aspirin, the overwhelming recommendation is if patients have other indications for these agents in their child PUA and B, you should treat your patient and they may drive an additional benefit. But certainly we need more prospective data. So my key takeaways are that portal hypertension is a treatment target, and SBBs are treatment for portal hypertension to prevent decompensation, largely with ascites, curvatilol is preferred. We have promising preclinical data, but our areas of unmet needs are validation of NILDA in diverse populations over time and see how that changes with treatment or even if we need to do that. We need to understand more about the clinical effectiveness and harms and discontinuation of NSBBs and tolerated doses, especially in our older populations. We need prospective human RCTs. We actually have very few. And then we need to think about additional agents that you'll hear about from Dr. Simonetto and some additional mechanisms that we should study. Thank you. Great. Next, we'd like to welcome Dr. Atusa Revi, who's gonna give us a lecture on predictors of outcome in patients with portal hypertension. You all in the back who are all standing, and there are a few seats up here. So if you wanna sit down, come on up front. Good morning, and thank you for the invitation. This is good about going third because a lot of my talk has already been covered. So my task is to talk to you about predictors of outcome in patients with portal hypertension. I don't have any relevant financial disclosure to this talk. So I want to start by, again, discussing the stages of advanced chronic liver disease. First, talk about predictors of outcome in those who are compensated, and then move on to those who are decompensated. So this, you have seen now multiple times. Advanced chronic liver disease is truly a spectrum. As you can see in the compensated stage, patients could have mild portal hypertension, CSPH, and then as the disease progresses, you have the first decompensation, and then continues to further decompensated stage. So if you look at the risk of death, you see that the most important transition in a patient that you're just diagnosing with compensated advanced chronic liver disease is that transition between compensated and decompensated stage. And the risk of death doesn't really take off until the patient gets to that CSPH and decompensated stage. So we do have, as was elegantly done by the previous speakers, we do have predictors of outcome in the compensated advanced chronic liver disease stage. The most important one is HPPG measured at 10 and above. So this study from Dr. Ripple shows that for those patients who have HPPG below 10, there's 90% chance of not developing decompensation, and then after that, when it goes above 10 millimeter, for each one millimeter mercury increase, there's 11% higher risk of developing decompensation. So we know that measurement of HPPG is not widely available. It's operator dependent, even if you can measure it, and not all sites are experts in doing so. However, we discussed that there are non-invasive tools that we can use in detecting CSPH as an important predictor of mortality. So this again has been discussed. We know that liver stiffness below 15 and platelet count above 150 is a very good marker to rule out CSPH and on the other end you have liver stiffness above 25 or other a little bit more grayer area of liver stiffness between 20, 25 and platelet count below 150. That could rule in CSPH. However, we do know that ideology matters. As was mentioned, liver stiffness above 25 has a pretty good positive predictive value of close to 90% in those who have viral disease or alcohol related liver disease or lean mash. When you look at obese mash group, this, that positive predictive value drops to about 63%. So not only with liver stiffness, we can predict presence or absence of CSPH but across the continuum, we can also predict risk of decompensation as well as risk of HCC. So this was a beautiful study by Dr. Rowe. Over 3,000 patients, median follow up about three years and as you can see, the risk of decompensation across the continuum as the liver stiffness increases from below 15 to 15 to 25 to above 25, the risk of decompensation gradually increases as well as the risk of development of HCC. Now this was in all comers and was not ideology specific. So as I mentioned before, ideology truly matters. So this is a beautiful graph that was, the high resolution of it was a courtesy of Dr. Thiele, recently published. So they looked at and basically put together two different studies looking at patients who have MASLD and those who have alcohol related liver disease. So you will see that the cutoffs for transient elastography are close but the risk of decompensation at each cutoff was much lower in those who had MASLD. So first let's look at those who had alcohol related liver disease. You can see with transient elastography below 10 kilopascal, the risk was about 3%. Then it went up to about 20. And then for those above 15, the risk was closer to 50% decompensation. If you look at those with MASLD, at about the same cutoffs, the risk was much lower. So you can see for transient elastography below eight and then eight to 12 and then lastly above 12. So looking at these two patient population, we see that at each cutoff, the risk of decompensation is much lower in patients who have etiology diagnosis of MASLD. Overall, in the group, in patients with diagnosis of MASLD who were followed for that median four years, 11% decompensated, moved to the decompensated stage. However, in a similar cohort of patients who had advanced, who had alcohol related liver disease, the rate of decompensation was fourfold higher. So it is important to realize that all of these markers in a non-invasive test that we use, it should be taken into account what etiology we're talking about. Now, not only etiology of the liver disease matters, but also type of decompensation matters. So this was a study that looked at about 500 patients moving from different stages of disease. So going from compensated without presence of clinically significant poor hypertension, then moving on to compensated with CSPH and then decompensation, which was defined as hepatic encephalopathy, ascites, bleeding, jaundice, and then outcomes of liver transplantation or death. So interestingly, those patients who presented with bleeding without other complications had a five-year mortality rate of 20%, which was significantly lower than those who presented with other type of decompensation, such as hepatic encephalopathy or ascites. So now let's move on from the compensated stage to decompensated stage. So we talked a lot about what are the predictors when you're in the compensated stage. Now, when you move on to the decompensated stage, we know that most patients decompensate in a non-acute manner. The most common decompensation is ascites. The PREDICT study looked at those who are admitted with their first acute decompensation, and they looked at them for about 90 days with outcome of a one-year mortality. So they identified three different clinical scenarios. So those who were stable acute decompensation, that was actually the most common scenario. So these were patients who got admitted, discharged, did not get readmitted, did not develop ACLF in the follow-up. Those who had unstable acute decompensation, so these patients had one or more readmission within the 90-day follow-up. And those who were in the PRE-ACLF group, which then went on to develop ACLF in the follow-up in the 90 days. So these three different clinical scenarios that I just touched on, not only they had very different one-year mortality, you can appreciate it here on the right-hand side that the PRE-ACLF group mortality at one year was about 60%. When they looked at the unstable group, it was about half of that, and then it was around 10% for the stable decompensated group. But also markers of inflammation was significantly different in these three groups. So if you look at this sort of left-hand graph, it shows you that the PRE-ACLF group, the group that then went on to develop ACLF, had significantly higher markers of inflammation, which was in this study defined as white blood cell and CRP. However, the group that had more surrogates of portal hypertension was unstable decompensated group, which was significantly much higher compared to PRE-ACLF or stable decompensated. So there are three different clinical scenarios that you can kind of take away. And in each one of them, there is a different pathophysiology to sort of predict mortality of one year. Now, once the patient decompensates, there's less known about further decompensation. So this was a really nice study that looked at patients who presented with grade two or three ascites, and as you can see, a little over half of them during follow-up developed further decompensation. And the further decompensations that I have listed here are in the order of being common. So refractory ascites was the most common, followed by hepatic encephalopathy, SPP, HRS, AKI, and variceal bleed was the least common. When they looked at one year mortality in those patients who presented with grade two ascites, MELD remained a significant predictor of mortality. Apologize, that's MELD and not MELD-sodium. MELD below 15 had a one-year mortality of three to 6%, and a MELD 15 and above had a 14% mortality at one year. The overall, in the grade three ascites, overall mortality rate at one year was 15 to 20%. So in this study, as far as looking at predictors of outcome in those who go into the further decompensated stage, they showed that in the overall cohort, in all of the patients who presented with grade two and three ascites, age and albumin remain significant as far as predictors of outcome. However, when they looked specifically at grade two ascites, age, albumin, as well as MELD remained significant. So with that, I wanna sort of summarize my talk that the advanced chronic liver disease is a spectrum. It is important to realize that the most important transition is between going from compensated stage to decompensated stage. The predictors of outcome are different, whereas the degree of portal hypertension is important and is an important predictor in the compensated stage. And we have non-invasive tools of measuring that. When you move to the decompensated stage, presence of inflammation, circulatory dysfunction, organ dysfunction measured by albumin MELD is more important in predicting patient's outcome. And with that, I conclude my talk. Thank you. Okay, I'd like to welcome the final speaker for this session, Doug Simonetto, who's gonna give us a lecture titled Regression of Portal Hypertension, Mechanisms and Therapeutic Strategies. And I also just wanna remind everybody that we're gonna have a Q&A session with all the speakers at the conclusion of this talk. There you go. Good morning. Thank you, everyone, for joining our session. So this is the last talk on portal hypertension. These are my disclosures. And so in this talk, we'll review portal pressure reduction as measured by HVPG as a therapeutic target in the short term, which is quite important as we develop new therapies for portal hypertension before you commit to long term trials with clinical endpoints, which may take several years to develop, as you heard earlier, we need to focus on short term benefit with potential new drugs. We'll discuss briefly the outcomes of ideology directed versus targeted therapies. And we spend some time learning about what's on the horizon with ongoing or upcoming phase two and three trials. But first, just as a refresher, remember, portal hypertension is driven by two main mechanisms. First, we have an increase in intrahepatic vascular resistance, 70% of which is driven by mechanical forces. So essentially fibrosis causing extrinsic compression of the sinusoids, and 30% of which is driven by a dynamic or functional component, which is driven by sinusoidal vasoconstriction, typically driven by a decrease in nitric oxide and increase in a vasoconstrictor called endothelin. The next we have increased portal blood flow, and that's driven by splenic arterial vasodilation. So we have an increase in the portal inflow in the portal vein that will further exacerbate the portal hypertension. And the splenic arterial vasodilation is thought to be caused by systemic inflammation driven by translocation of bacterial products from the gut. So as we discussed earlier, so we're gonna focus on therapies that can reduce hepatic venous pressure gradient in the short term. And again, that's important so we can assess whether drugs are effective before we commit to long-term trials focusing on clinical endpoints such as first decompensation. So with that, it's important for us to understand what are the thresholds for HVPG that will translate into better clinical outcomes? So we've known what they are for quite some time now, and those are HVPG reduction by at least 10%, ideally 20% from baseline, or if we can reach HVPG equal or below 12. And we know that reaching those thresholds will translate in better clinical outcomes. In this old study here, we can see the patients that were HVPG responders had significantly lower risk of variceal bleeding at five years compared to non-responders. In the same study, we also showed that these patients that responded had lower incidence of hepatic encephalopathy, and even more importantly, lower mortality at five years. So I know Dr. Serpic covered beta blockers pretty well. I just wanted to show this slide for you all as a reference so we have something to compare the new drugs with something that we currently have available now. As was alluded to before, curvidilol is the most potent non-selective beta blocker because it also has the anti-alpha-1 effect. And you can see that it can lead to a greater reduction in HVPG of about 20 to 23% compared to 11 to 13% with our traditional non-selective beta blockers. In addition, up to 75% of patients on curvidilol will achieve the HVPG response rate compared to up to 50% with, again, traditional non-selective beta blockers. So this is something for you to remember as you have something to compare against. So now as we move on to strategies for portal hypertension regression, basically two mechanisms. So portal hypertension regression can happen spontaneously just by removing the ideology of chronic liver disease, or we can target specific mechanisms along the pathway I mentioned earlier that can reduce portal pressure. Spontaneous regression of portal pressure has been long shown, and this is a study from 1960 New England Journal of Medicine where they found that patients with alcohol-related cirrhosis had a significant decrease in HVPG just with sustained alcohol abstinence. And since then, several other studies have confirmed this finding. So with sustained alcohol abstinence, we can reach up to 40% average decrease in HVPG from baseline with up to 60% of patients achieving HVPG response. With hepatitis C eradication, we can reach up to 30% average decrease in HVPG from baseline with anywhere from 20 to 100% of patients reaching the main endpoint of HVPG reduction. Vital hepatitis B suppression, average 20% reduction from baseline, up to 40% of patients achieving HVPG response. And finally, with lifestyle interventions, combination of nutrition, physical activity has been recently looked at, and they found that 20% average reduction on HVPG from baseline and up to 40% of those patients reaching, again, the significant outcome on HVPG response. Now, one caveat with all these studies is that they were all very short term, anywhere from eight weeks to six months. So you may argue that it probably will take a little longer for protopressure to decrease if the underlying mechanism here is fibrosis regression. So fibrosis won't regress quickly, so it may take time. So if it went long enough, maybe the protopressure will continue to decrease over time. So this one study looked at that question, it had two-year follow-up, and they enrolled patients with hepatitis C cirrhosis with CSPH, so all patients had CSPH at baseline. And you can see that at two years, unfortunately, still a little over half the patients had CSPH, with over a third of the patients still having HVPG equal or greater than 12. And again, you may argue, okay, if you wait a little longer, you're gonna see a progressive decrease in HVPG. However, I would argue that the longer it takes for us to reach those thresholds, the longer our patients are at risk for portal hypertension-related complications. So with that in mind, we need to continue to develop and investigate new therapies that can target portal hypertension itself, so not just rely on spontaneous regression. So they can be targeting the intrahepatic vascular resistance, either the structural component or the functional. Drugs that target a structural component are basically our antifibrotic drugs. And here's a list of different classes that have shown great promise in preclinical studies. Unfortunately, they haven't been translated yet into human trials, except for some data that Dr. Serper mentioned on anticoagulants. And then we have drugs that target the functional component of the intrahepatic vascular resistance. So these drugs basically promote sinusoidal vasodilation by either increasing nitric oxide in the sinusoids or decreasing endothelin. And you can see Covetilol is the only beta blocker listed here because it has the anti-alpha-1 effect. Then we have drugs that can target the increased portal blood flow. So they are basically splanchnic vasoconstrictors, either directly or indirectly. So example, rifaximin potentially probiotics can reduce splanchnic vasodilation by decreasing the translocation of bacterial products from the gut. And then we have direct vasoconstrictors, two of them available to us. They are short-acting, octreotide and telepressin, both given intravenously. Of note, there is a subcutaneous form of octreotide that's long-acting. It has been trialed in patients with portal hypertension without benefits. So octreotide only has a role in the acute setting for variceal bleeding. And then we have long-acting vasoconstrictors. The main ones that we have available to us, the non-selective beta blockers. So note that statins are on both columns here. So statins can decrease the intrahepatic vascular resistance by increasing nitric oxide in the sinusoids, but it can also promote splanchnic vasoconstriction because it has its own inflammatory and antioxidant effects. So finally, what do we have on the horizon? So we have a few phase three trials ongoing. So we have this Roxaban, it's a European trial investigating rivaroxaban in patients with CSPH. And essentially, the primary endpoint for that study is time to transplant-free or decompensation-free survival. And the study, I understand, has stopped enrollment. We haven't seen the results yet. We have the RAISE trial, which is investigating rifaximin compared to placebo in patients with zofagenoviruses. The primary endpoint here is the HBPG reduction at eight weeks. And then finally, we have four clinical trials with the statins, two of them investigating rosuvastatin and two using sinvastatin, including the SACRE trial, which is a trial done by the VA system here in the U.S. They are looking at sinvastatin, 40 milligrams a day with the primary in patients with CSPH, with the primary endpoint being decompensation-free survival. And then finally, the LiverHope is investigating sinvastatin and rifaximin. The opinion trial is enrolling patients with decompensated cirrhosis, and the primary endpoint is incidence of ACLF. And then finally, we have a few phase two trials, which I will discuss in a little more detail here. So first is belapectin. So belapectin is a galactin-3 inhibitor. So galactin-3 is a protein released by macrophages that can activate stellate cells and promote fibrogenesis. So belactin can inhibit galactin-3. So a recent study has been published looking at belactin compared to placebo. And unfortunately, that trial, they did not reach the primary endpoint, which was HVPG reduction at one year. However, in a subgroup analysis, they found that patients that received low-dose belapectin had a significantly greater decrease in HVPG compared to placebo. And the authors also found that no patients in the low-dose belapectin arm developed varices at the end of their study at one year compared to 18% of patients that received isophageal varices at the end of the study in one year compared to 18% of patients that received placebo. So this results served as the basis for the NAVIGATE trial, which will be investigating Bella packing at two different doses compared to placebo, with the primary endpoint being development of isophageal varices. In this study, they are only enrolling patients with MESH, cirrhosis, and any portal hypertension, including metric oxide pathway. Metric oxide, as I mentioned, is a potent vasodilator that's decreased in the sinusoids in the setting of cirrhosis. And we have two drugs showing promise here. So one is the 5-methyltetrahydrofolate. A study was just published this past month looking at 5-MTHF when added to propranolol. It led to a significant reduction in HVPG compared to propranolol alone. And 5-MTHF prevents degradation of an important cofactor for ENOS that makes nitric oxide. So we'll probably hear more about this drug in the near future. And then we have a new compound that doesn't have a name yet. It's BI-685509, which is a direct activator of the soluble guanylucyclase. So you can activate this enzyme that produces cyclic CMP without the need for nitric oxide being present. So recently just published on hepatology communications was a paper phase one trial that showed that this compound is safe in patients with cirrhosis. It was a phase one safety trial. And now we have two upcoming trials with this compound. One is placebo-controlled in patients with alcohol-related liver disease. The other is an open-label single-arm trial in patients with hepatitis B, hepatitis C, and MASH. And the primary endpoint for these trials is HVPG reduction at 24 weeks for the first trial and eight weeks for the second. And finally, endothelin, as I mentioned, is a potent vasoconstrictor that's increased in the sinusoids. So in this study, the authors injected a selective endothelin A antagonist in the hepatic circulation, and they found a significant reduction in HVPG within 30 minutes. And when given orally, this drug also reduced HVPG, although not as much. And the ZIL trial is a phase two A to B study. They will be investigating zybotentin, which is a selective endothelin A antagonist, and dapaglifazine, which is an SGLT2 inhibitor. There are two parts. Part A was just completed. Part B will have five arms and three doses of zybotentin. And the primary endpoint for this study is HVPG reduction by 20% from baseline or reaching levels below 12 at six weeks. So this is my summary slide, and I'll end with key areas for future research. As I mentioned earlier, we need to identify and validate noninvasive modalities that can predict or estimate HVPG reduction noninvasively that can be used in clinical trials, but also in clinical practice. We also need to identify and validate thresholds after which screening, endoscopy, or even beta blockers can be stopped. And finally, there's a lot of excitement with statins, and we're all looking forward to the results. So I'd like to invite our four speakers up, and we're going to have a Q&A. Anybody who's interested in asking a question, please come to the microphone and identify yourself before asking the question. I have a question. Well, there is a population that is very challenging to evaluate with a noninvasive test, that is the patients with morbid obesity. And patients with morbid obesity, you know, you may not be able to do a transient elastography by fibroscan. Some of these patients are considered for bariatric surgery. Some of these patients have normal platelet count, no imaging findings of portal hypertension. So what would be the approach to evaluate in a cost-effective way these patients, you know, just doing MR elastography, doing invasive measurement of portal pressures, or just an operant endoscopy and try to rule out the nurses and then proceed with bariatric surgery? I think you can do MRE. You could perhaps also do spleen stiffness. The issue with spleen stiffness with magnetic resonance is not as known or done at some other places. I mean, with Dawn and some of the discussions that we've had for the NILDA group, one of the issues is that MRE is not widely available. So spleen stiffness from MRI, it's even less available than that. But you have to consider that in some of the patients that have grade 3 obesity, the transducer for the mechanical vibration might not do a good job, and even MRE or spleen might not be reliable. So you likely are going to have to do a biopsy on those patients or an HPPG. Thank you. Do you want to go to the back? Hi, this is Giorgio Anu from the University of Washington, and I wanted to congratulate the speaker's excellent presentations. I think we underutilized non-selective beta blockers and maybe even more carvedilol, but if I could be devil's advocate and say perhaps that most of the benefit in the clinical trials and in the observational studies is derived from reduction in ascites. And so why not maybe just alternatively wait for the ascites to happen and then treat it instead of treating 100 people to prevent 10 ascites incidents from occurring? And then the corollary to that is do you think the portal hypertension is actually causing damage in the liver? Is it a physiologically problematic event that's destroying the liver, or is it just an effect of damage? Is it cause or effect? So everybody's focusing on treating the portal hypertension, but is it just an effect or is it cause? And then, again, just lastly related to the last one, we have a very effective treatment for portal hypertension, which is TIPS, and you didn't mention it, so why not just do it TIPS if you think portal hypertension damages the liver, just do it TIPS, you've solved the problem. Well, these are hard questions, George. I'll actually think about this a lot where we know that portal hypertension begets portal hypertension, right? The worse it is. So is it actually an outcome, or is it a predictor? I'm actually not convinced myself that portal hypertension is a predictor of decompensation versus a surrogate marker of something else. That's a really good point, and I don't know that there's a really good answer for that. I can tell you, on the beta blocker front, we do need a lot more real-world data because even anecdotally, and if you talk to people who treat a lot of patients with cirrhosis, beta blockers are not that well-tolerated, and even Carvedol at 6.25 milligrams is not necessarily well-tolerated by individuals who are older. It does drop your blood pressure, it makes you dizzy, so we need to understand better what these beta blockers do, and I think what was your other point? I think you had multiple. The TIPS. I guess the cause and effect, you could solve that answer by TIPS if you thought cause or effect. It was interesting, in the VILA study with enoxaparin, even though you saw a drop off over time, with anticoagulation, you saw no PVT and no decompensation in the enoxaparin group. Would patients with CSPH be willing to undergo a preemptive TIPS? I don't know. I think that could be studied. I think there is evidence actually showing that NSBBs don't necessarily prevent the formation of varices, so what would happen with a TIPS? I think that needs further study. George, I don't want to steal the speaker's thunder here, but there are data, there are cell data on endothelial cells and stellate cells that show that pressure and flow have adverse effects at the cellular level, so I think it's, I've thought about this for 25 years, so it's a great issue and something that we really need to tackle. Sorry about that, you guys. Does Josh Levitsky from Northwestern just echo George? These were fantastic talks. I'm going to ask another challenging question, which is, none of you talked about prevention of portal hypertension, and I know the data probably aren't real supportive, but I'm just curious your thoughts. If I was a patient with cirrhosis and didn't have portal hypertension, why shouldn't I put myself on a low-dose beta blocker or baby aspirin and a statin to prevent portal hypertension, also with taking care of whatever is causing it? Any data for doing something like that, or are there any studies that are looking at the prevention of getting portal hypertension in patients with cirrhosis or even like MASH with F3, where they clearly data has shown that a lot of them have portal hypertension or are at risk for it? Go ahead. Just a brief comment. You know, tackling the etiology of liver disease, I think that's going to be the most important thing. And in the case of NASH, or MASH, the important thing is going to be to go into those lifestyle factors, like changing the lifestyle of the patients, and very importantly also comorbidities, because comorbidities have a lot of effect of what eventually happens to the progression of the liver and the occurrence of outcomes. And with that being said, weight loss and exercise are actually outstanding interventions for decreasing portal pressure. Just being realistic of our patient population. It has to be studied. Someone who does a lot of clinical research, it has to be studied, so there are a lot of mechanistic studies that are elegant, a lot of observational data we can generate, but you have to study primary prevention strategies. Thanks. Back. Hello, this is Angelo Zambudimato from Brazil. I would like to ask for Dr. Serper and Dr. Cimure about the way they verify if patients are responders to curvidilol. Do you already think we could use the change in values of stiffness? How do you do it in clinical practice? Thank you, Angelo, for a great question. So I don't think we have data yet. That's one area that needs further research. How can the new non-invasive tests will perform at predicting HVPG reduction? So in clinical practice, we wait for the clinical endpoints and clinical outcomes. We don't routinely measure, it's not in my practice, HVPG, so that studies have shown HVPG reduction in all clinical trials, and that's clearly what we have as the gold standard for. In the real world, we start patients with beta blockers and assume that they will be responders. With the old traditional beta blockers, we target the heart rate as a surrogate for that. With curvidilol, it's a fixed dose, and most patients will be responders, but not all of them, as you saw in the data. What do you think about splint stiffness variation? Right, so again, they're already in further research, and I think it's important for us to look at that for potential improvement and resolution. Thank you. Keqing Hu from the University of California, Irvine. Thanks for the panel. Great talk, great presentations. My questions relate to the safety of studying with cirrhosis. We'll talk about the benefit and now potential. I'm concerned, people with cirrhosis, are they okay to take studying? What dose? What regimen? So many options, too. I want to ask the panel the question, your opinion, while we're waiting for trial data, what's your opinion about using studying in people with cirrhosis, sometimes not for prevention, for potential improvement, just for muscle death? Yeah, so I can comment on that. I removed that slide because of the short time I had. So there's an old study on hepatology that looked at fluvastatin compared to placebo. The primary endpoint was safety, and they showed that fluvastatin did not lead to an increase in AST and ALT compared to placebo. In fact, the placebo group had slightly higher elevation of liver enzymes compared to fluvastatin. And more recently, the Liver Hope Safety Trial was published that looked at safety of sivastatin, 20 or 40 milligrams a day, and those were patients with decompensated cirrhosis already. So patients that got a 40-milligram dose actually had a higher risk of rhabdomyolysis and muscle toxicity, but the 20-milligram dose of sivastatin appeared to be safe. And that's what's being studied now on the efficacy liver hope study. Thank you. Hello, this is Dr. Akash from Calcutta, India. In the last guidelines and guidelines, mere presence of portal hypertensive gastropathy or endoscopy has been considered as a surrogate for CSPH. Our studies, mostly on noninvasive predictions, have been based on high-risk varices, some on presence of any varices. As we move on of using carvidolol for non-varicial uses, could we change the goalposts of using noninvasive detection of any presence of CSPH, which might be even marked by portal hypertensive gastropathy, to make endpoints for studies? The second one is, with the use of rivaroxaban, do we go into a zone of using parameters like thromboelastography and VWF to stratify patients who might benefit from long-term anticoagulant therapy? I'm going to tackle the first one. On portal hypertensive gastropathy, and I actually believe Don just published that on its correlation with HPPG, most patients will have clinical significant portal hypertension, but not all of them. The clinical trial, or the studies that have been done, have included either high-risk varices, large, small, with high-risk astigmata, or any varices. I think PHG has been really not included in most of these studies. We tend to disregard that. I believe that in clinical practice, and especially because most of our patients have muscle, or have features, that obesity, maybe, even if they don't have steatosis, that we just have to consider any of them as an equivalent to clinical significant portal hypertension. That's what I do in my practice. In terms of apixaban, I think there's insufficient evidence to do anything like riveroxaban to measure anything like anti-factor 10A levels or anything. I do want to mention just one thing. It had been noted, both in the cardiology literature and a little bit in our literature, that actually apixaban is associated with less bleeding than riveroxaban, so even though this is completely unofficial, I would favor apixaban if you're going to use it in cirrhosis, child A and B. I think we have time for one additional question. Jeff Massoumi from Northwell Health in New York. Excellent talks. We learned that mazzled patients decompensate at lower portal hypertension level. We know that majority of them have diabetes. We know that majority of the diabetic patients have some elements of diastolic dysfunction and CKD, which is not easily, basically, easy to be evaluated. At what point you would look at cardiovascular and kidney modifying factors in order to control the decompensation like SGL2 inhibitors? I think that anything you can do, I don't have any evidence for what I'm going to say, but I think it's pretty clear on or there's direct evidence in the literature to say that whenever you improve comorbidities, you also improve liver-related outcomes. That's particularly true for the patients with NASH or in the older literature. I don't have anything to directly answer your question, but I think that whatever we do to improve cardiac outcomes and kidney outcomes or improve or prevent anything that will be cirrhotic cardiomyopathy or HRS will definitely improve the portal comprehensive behavior. Have you looked at your VA data on SGL2 inhibitors? Just them alone? Well, we can ask Dr. Nadeem Mahmood and I'll ask him, but he actually has a bariatric surgery cohort so he's actually looking at pharmacotherapy. We do have one study SGLTI and I don't think we have a firm idea of what it does with decompensation, but we can further study it. Yeah. I also want to say that I think it is important to think about it as chicken or egg, right? So yes, the muscle population do have cardiac diastolic dysfunction, lots of CKD, but also there is diastolic dysfunction and kidney injury happening as circulatory dysfunction as the patient decompensates. So it is important which stage we're talking about, about removing that cause and effect. Doug, did you have a question? Yeah, I was just going to add with SGLTI too, I'm very excited about this class of drugs. I remember we have seen already in the Lancet just published that it may actually prevent progression of CKD. So a study design of CKD patients, so very exciting data. Okay. Unfortunately, we're out of time and really sorry that we couldn't get to the other questions, but I think that just speaks to the interest of the crowd. So I'd like to thank the speakers for awesome talks. You guys did a great job. Thank you very much. Applause One of the best sessions I've seen in recent memory. So we're going to break for 12 minutes now. So for the portal hypertension fans, please come back at 10. If you want to suggest topics for future SIGs, please come back. It's an open session. So we'll reconvene at 10. Thanks very much everybody.

portal hypertension

assessing

HPPG measurements

non-invasive tests

treatment options

beta blockers

anticoagulants

ongoing research

therapeutic strategies

TIPS

cirrhosis