Well, welcome, everyone. It's a treat to join you in this virtual format to discuss a very important topic to all hepatologists and gastroenterologists, and that is treatment strategies beyond first-line therapy for autoimmune hepatitis. I'm delighted that we have two excellent speakers who are personally involved in the care of these patients and in research in the area to provide these two lectures. The first speaker will be Aparna Goel, who is Assistant Professor of Medicine in the Division of Gastroenterology and Hepatology at Stanford University. She received her M.D. degree from Thomas Jefferson and completed residency in internal medicine at UCSF, followed by fellowships in gastroenterology and transplant hepatology at Mount Sinai in New York. She practices general and transplant hepatology. Her clinical and research interests include management of patients with autoimmune hepatitis, primary biliary cholangitis, and primary sclerosing cholangitis. Our second speaker will be Therese Bitterman, who goes by Tess. Tess Bitterman is an Assistant Professor of Medicine in the Division of Gastroenterology and Hepatology at the University of Pennsylvania. She's also a faculty member in the Department of Biostatistics, Epidemiology, and Informatics, and a senior fellow at the Leonard Davis Healthcare Economics Institute at Penn. She is an elected member of the AASLD Clinical Research Committee, and she is about to join the editorial board of our journal, Liver Transplantation, and serve on the medical advisory board of the Autoimmune Hepatitis Association. She works clinically as a transplant hepatologist, and her research interests focus on the long-term consequences of chronic immunosuppression and heterogeneity in the management of liver transplant recipients and patients with AIH. She received the 2020 Autoimmune Liver Diseases Pilot Research Award to establish the epidemiology of AIH in the United States and to evaluate the use of evidence-based management strategies in a national cohort of patients with AIH. She is also the recipient of an NIDDK K08 Career Development Award to study practice variability in immunosuppression management among liver transplant recipients and its impact on outcomes. So as you can see, I'm delighted to have such talented speakers who are truly rising stars in the constellation of hepatology provide these lectures. So thank you for the kind introduction. I'm going to just go through a brief outline of what we're going to talk about today. I'm going to review the most updated guidelines from the AASLD on first-line therapy, discuss toxicity from first-line therapy, and potential alternatives to first-line therapy. And then my colleague, Dr. Goyal, is going to provide an in-depth review of second-line therapy as well as efficacy and safety considerations. So our first case is a 68-year-old woman with a history of hypothyroidism and diabetes who has fatigue and nausea for one month. Blood work by her primary care physician showed elevated transaminases with a normal cholestatic profile. She is admitted for further workup and is found to have elevated autoimmune markers with a positive ASMA elevated IgG. And her anti-mitochondrial antibody is negative. She has negative viral studies and no recent drug exposures. She subsequently has a liver biopsy that shows chronic hepatitis, portal inflammation with lymphoplasmicytic infiltration, interface hepatitis, and no biliary changes. Ultimately diagnosed with autoimmune hepatitis based on a simplified score of seven of eight. I want to walk you through the most updated guidelines for first-line treatment of AIH. The first distinction that the updated guidelines provide is to think about our patients into three different groups. AIH, AIH with cirrhosis, and acute severe AIH. As a reminder, acute severe AIH is AIH with jaundice and elevated INR and no encephalopathy. The second part of the algorithm distinguishes induction and maintenance phases of treatment. The goal of induction is to achieve biochemical remission and that is defined as an AST, ALT, IgG less than the upper limit of normal. As a reminder, some of the prior definitions actually use less than two times the upper limit of normal. And that becomes important when thinking about older studies and treatment endpoints in those studies. The first component of induction treatment for AIH is corticosteroids. As a reminder, if a patient does not have cirrhosis or acute severe AIH, prednisone or budesonide can be used. If cirrhosis or severe AIH are present, no budesonide should be used. And if acute severe AIH is present, patients should be treated with high-dose prednisone or IV steroids. The second component is azathioprine. In this updated guideline, they note the importance of checking a thiopurine S-methyltransferase or TPMT activity. As a brief reminder of azametabolism, azathioprine is a thiopurine prodrug. It is converted to 6-mercaptopurine or 6-MP and ultimately forms thioguanine nucleotide or 6-TGN, which is the active metabolite. Complete TPMT deficiency occurs in about 0.3% of the population, what happens here is that in the absence of TPMT activity, increased shunting of azathioprine to the active metabolite occurs and this causes severe myelosuppression. Going back to the algorithm, once TPMT has been checked, then azathioprine can be started in patients with or without cirrhosis, typically two weeks after steroid use. In acute severe AIH, the consideration is to start azathioprine after the cholestasis has resolved. Back to our patient, she starts on prednisone and azathioprine, however a month later, she calls that she's been admitted to the ER with abdominal pain, nausea and vomiting. She's found to have elevated amylase and lipase. A CAT scan shows mild fat surrounding around the pancreas and she's diagnosed with acute uncomplicated pancreatitis. She's treated supportively and her aza is discontinued. She's referred now to us for further management of her AIH. A few brief words about aza-related pancreatitis. First reported in the early 1970s, this is believed to be an idiosyncratic drug reaction, typically occurring within weeks to months of use. Aza is considered a definite cause of drug-induced pancreatitis and the true incidence, while unknown, is estimated at less than 5%. The mechanism remains unclear, but possible predisposing risk factors have been studied, including genetic factors, such as xanthine oxidase polymorphisms and HLA variants, as well as female gender and extremes of age, and there has been this long debated relationship with inflammatory bowel disease, particularly Crohn's disease. Things to note about idiosyncratic adverse events from aza. Number one, they occur early, often less than a month after initiation. In addition, they are more common than late dose-dependent issues. And lastly, they are the most frequent reason for drug discontinuation. Our second case is a 62-year-old male with prostate cancer who is admitted to an outside hospital with abnormal liver enzymes. He's ultimately diagnosed with autoimmune hepatitis based on serologies and liver biopsy and started on prednisone 40 milligrams daily. He then sees us in outpatient GI follow-up and starts azathioprine 100 milligrams daily. However, three weeks later, he presents to a local hospital with two to three days of fatigue and fever up to 104 Fahrenheit. He's tachycardic with borderline low blood pressure, normal lactic acid, and normal white cell count. Complete infectious workup was negative, including a pan-CT scan. However, he continues to have daily cyclical fevers up to 104 with shaking chills. His azathioprine is discontinued with resolution of the fevers, and he's transitioned to mycophenolate, mofetal, or MMF as an outpatient. Other idiosyncratic adverse events from azathioprine include fever without leukopenia, like in this prior case. This is believed to occur in less than 5% and is thought to be a hypersensitivity-type reaction. The main culprit is thought to be an inactive metabolite that is produced early in this metabolism of azathioprine. We know this because it does not occur if 6MP is given. It can occur after days of use and can be associated with other symptoms such as rash, nausea, vomiting, diarrhea, and increased liver enzymes. And notably, the symptoms recur and can be even more severe if re-challenged. Other idiosyncratic adverse events from AZA include nausea and vomiting up to 20% in some series and a rare acute cholestatic liver injury in 0.1%, in which patients present with fatigue and jaundice about 2 to 12 months after initiation. These patients will have a mild to moderate mixed liver injury, a liver biopsy that shows bland cholestasis with minimal inflammation. Typically, this resolves with discontinuation, though can progress to vanishing duct syndrome. Dose-dependent AZA toxicity is going to be discussed next. Pharmacologic issues include TPMT deficiency, which we already discussed in part. This can be complete or partial. Another important issue is that of major drug interaction with xanthine oxidase inhibitors, which are over here in the pathway. These drugs include drugs for gout, allopurinol, and fibuxostat. Similar to TPMT, if one is to block xanthine oxidase, there is increased shunting of azathioprine to the active metabolite 6TGN and severe myelosuppression. Other dose-dependent AZA toxicity includes myelosuppression in the absence of a TPMT-related issue, which can occur in up to 25% of patients with leukopenia and thrombocytopenia in 5%. It's important to note that this myelosuppression effect can be increased in patients with cirrhosis. Hepatotoxicity can also be manifested by mild transaminase elevations in the first 12 weeks. These typically are self-limited and resolve with holding or decreasing the dose. And sometimes there may be a role for checking AZA metabolites, such as 6TGN and 6-methylmercaptopurine, to help identify drug toxicity versus non-adherence and non-response. Other hepatic issues to be aware of with azathioprine in the long term include non-cirrhotic portal hypertension and nodular regenerative hyperplasia, which typically occurs several years after use, and hepatotasplenic T-cell lymphoma, which is very rare. In this recent study, they looked at reasons for discontinuation of azathioprine in the first year and found that this is actually a quite common event, occurring in about 10 to 20% of patients with AIH. And this was mostly due to side effects, including GI complaints, hepatotoxicity, and pancreatitis. Notably, there was no difference if AZA was used early versus late and no difference by AZA dose. What about early toxicity from corticosteroids? Well, we know that the adverse effects of corticosteroids are increased in patients with cirrhosis, which is an important thing to keep in mind. About 50% of patients will discontinue steroids due to cosmetic changes, such as acne, fluid retention, or obesity. In this trial, comparing budesonide versus prednisone, about 50% experienced at least one side effect of prednisone, and that occurred in the first eight weeks of treatment. I show this mostly to note that the vast majority of these side effects were in fact these cosmetic or physical changes, such as moon phase, acne, and hirsutism. There are other very commonly appreciated early adverse events of corticosteroids, and those include hyperglycemia, increased blood pressure, insomnia, irritability, and dyspepsia. So when should one consider alternate first-line therapy? Well, the reality is that it's difficult to avoid corticosteroids. In part, this is because we know that AIH is typically very steroid responsive. As I've alluded to, budesonide is an established first-line therapy to induce remission. A potential benefit of budesonide is its 90% first-pass metabolism, which has the potential ability to reduce side effects. In that one trial that I had mentioned previously from Manns et al., they found that the side effects with budesonide were 28% versus 53% with prednisone induction. However, in other studies, this benefit in terms of adverse events with budesonide is a little bit less clear. The goal should be to minimize the duration of corticosteroid use, and there is potentially a role for transitioning from prednisone to budesonide during induction therapy if perhaps the patient is not tolerating prednisone therapy, as long as they don't have cirrhosis. In addition, it's important to consider these other preexisting conditions, not only the well-appreciated metabolic syndrome risk factors, such as diabetes and hypertension, but also glaucoma, cataracts, osteoporosis, and psychiatric history. What about corticosteroid-free induction? Well, early studies from the 1970s showed mortality rates of 20% to 30% using azathioprine monotherapy for induction, so unlikely to be a reliable option. However, there is actually a recent small series published from the Children's Hospital of Philadelphia, right next to where I practice, that suggested that azamonotherapy could be safe and effective in select children with mild AIH. In addition, there are several small studies from the 1990s using cyclosporine monotherapy to induce remission, and I'll talk about these a little bit later. Should one avoid azathioprine? Well, as I mentioned, complete TPMT deficiency is a contraindication to azathioprine use, but also preexisting cytopenias, hypersplenism, and prior aza intolerance may make one not have azathioprine as an option for first-line therapy for AIH. In addition, there are some common drug-drug interactions to think about. These are not necessarily reasons to avoid azathioprine altogether, but just to keep in mind that warfarin and angiotensin converting enzyme inhibitors, which are two very commonly used drugs, do have drug-drug interactions with azathioprine. Our third case is a 32-year-old woman with undifferentiated mixed connective tissue disease, elevated liver enzymes, elevated ANA, ASMA, and IgG. She is diagnosed with AIH and bridging fibrosis based on a liver biopsy and her serologies. Post-procedure, she developed abdominal pain and went to a local ER, and there she was found to be five to six weeks pregnant. This is, in fact, a true case that I met this woman. She had been biopsied in an outside facility where they, in fact, did not check a pregnancy test before her biopsy. She now presented to us for further management. A couple words on azathioprine and pregnancy and breastfeeding. In this one survey study in Scandinavia, they found that up to 50% of women with autoimmune hepatitis do not consult their doctors before pregnancy. This is important when we think about treatment. While azathioprine is technically Category D, many agree that it can be continued in pregnancy, and this is because there has been no sign of teratogenicity noted in studies of post-transplant women or women with inflammatory bowel disease who get pregnant. The other important aspect is that alternatives may have much worse pregnancy risk. For example, MMF, and I'll go into this a little bit later. We know that pregnancy is associated with decreased disease activity from an AIH standpoint. However, there is an increased risk of postpartum flare, and that is correlated with the degree of remission that had been achieved prior to pregnancy. With regards to breastfeeding, it is transmitted in small quantity through breast milk, but appears safe. Nevertheless, in that prior study from Scandinavia that I had mentioned, they did find that over 50% of patients of women stopped or reduced their azathioprine during pregnancy or breastfeeding. In terms of MMF as first-line therapy, there are three studies, and I'll walk through all three of them. The first is a retrospective single-center study of 29 patients who received MMF for autoimmune hepatitis. 16 patients received MMF with steroids. One received MMF alone as first-line, and the remainder actually received MMF as second-line therapy. They had a relatively high intolerance rate of 34%, but they did note a high rate of biochemical remission. I put an asterisk here as they defined biochemical remission was the older definition of less than two times the upper limit of normal. The big caveat, however, of this study was that, as I alluded to, there were patients included who were receiving MMF as second-line, and the study did not distinguish the outcomes of first-line versus second-line users. The second study is a prospective study of 59 consecutive treatment-naive patients who received MMF with prednisone. 88% had biochemical remission, i.e., transaminases less than the upper limit of normal, and most within three months. There were zero non-responders, and 3.4% had to discontinue MMF due to severe side effects. Things to keep in mind with this study, it is a single-center study, and there was no comparator group. And lastly, by the same authors as the prior study, there was an open-label observational study of 131 consecutive AIH patients receiving either MMF with prednisone, 109 patients, versus 22 with prednisone with or without azathioprine. Importantly, the patients who received prednisone with or without azathioprine were actually patients that did not consent to the original MMF protocol. So this was not a randomized study. They did find that the MMF group were more likely to have biochemical remission, 71% versus 45%. Important caveats here. It is a single-center study. As I mentioned, it was non-randomized. There is also in the study some limited characterization of the types of patients that were in this comparative group. And lastly, the outcomes of the comparative group are a little atypical. So MMF as first-line therapy is not yet a slam dunk. The data remain very limited versus current standard of care. There are no randomized clinical trials and no multi-center studies. However, the suspicion is that MMF is probably non-inferior to the standard of care. Potential issues with MMF remain. Number one, higher cost. This issue of not being able to use it in women who wish to conceive. As a reminder, according to the U.S. Food and Drug Administration, patients being started on MMF have to have a negative pregnancy test within a week and have to have two methods of birth control for four weeks prior and six weeks after use. And lastly, there are some adverse effects related to MMF such as GI toxicity and infections. The last case I'm going to present is a 24-year-old woman with lupus and elevated liver enzymes. She's diagnosed with AIH based on serologic and histologic criteria. Importantly, however, to this case, she is actually currently on azathioprine for lupus and has been so for two years. In addition, she's had prior severe GI intolerance to MMF requiring hospitalization. She has started on prednisone 60 mg daily with improvement in her liver enzymes and her provider has started her on tacrolimus twice a day, 1 mg, about 6-8 weeks after steroids and with a goal tacrolimus level of 5 ng per ml. The data on calcineurin inhibitors as first line therapy is presented here. With cyclosporine, there are actually no data in adults. There are several small studies in children of about 10-15 children of which a small group were treated as first line. These patients either received cyclosporine monotherapy for 6 months followed by low dose prednisone with AZA or cyclosporine in combination with azathioprine and MMF up front with or without prednisone. The authors of these studies note that the use of cyclosporine up front allowed for steroid minimization although some of the children did remain on long term steroids albeit at lower doses. These studies show that cyclosporine does appear efficacious in the long term but there are some side effects to be aware of hypertrichosis, gingival hyperplasia and kidney dysfunction. With regards to tacrolimus, there is actually only one study in adults using tacrolimus as first line therapy. This showed 70-80% had reductions in AST and ALT though again these were not the standard remission criteria. There is one small study of 20 children and young adults receiving tacrolimus as first line therapy. Important thing to note with that study is that the majority 14 of those 20 children needed low dose prednisone or azathioprine to achieve remission. Again, these studies were all observational with no clinical trials or comparators. Thank you, Dr. Bitterman, for sharing these interesting cases and highlighting the scenarios in which alternative first-line agents are indicated. We will now spend the rest of this webinar discussing the management of patients that do not respond to first-line therapy. So to review, there's four indications for second-line therapy. Dr. Bitterman has covered the first two, which are contraindications to first-line therapy or intolerance. We'll now review the use of second-line therapy for non-responders. Non-responders include patients that have an incomplete response to standard of care or treatment failure with standard of care therapy. Now, remember, as Dr. Bitterman mentioned, the goal of treatment is to achieve biochemical remission. The definition of biochemical remission is a complete normalization of your serum aminotransferase levels, an IgG, an elimination of portal inflammation and interface hepatitis histologically. The upper limits of normal that are defined for your ALT are less than 19 for women and less than 30 for men. A patient's individual time course to achieving biochemical remission can look like any of these graphs up above. So you can have a slow, steady decline in your ALT. You can have a more rapid decline in your ALT. You may have an initial decline, need to add second-line therapy, and then an eventual final decline to lower levels of upper-limited normal. The key is that this stringent definition of decreasing your ALT levels to upper limit of normal or below is very important in terms of preventing the progression in the need for liver transplantation. The older definitions of remission, as was alluded to in the 2002 guidelines and even prior to that, aim for ALT values that were around two times upper-limited normal and still allowed for ongoing portal-based inflammation. So when you look at the older data, about 75% of AI patients would meet the definition of remission, the more liberal definition of remission that allows for higher levels of ALT and ongoing portal-based inflammation. But when you follow those patients that, quote-unquote, achieve remission, about half continue to have progressive liver disease. So the reason why the definition was altered and revised to be more stringent was to prevent the progression of liver disease and the need for transplantation in this population. Now, with the more stringent definition of AIH and remission, about 25 to 50% of people will achieve remission with first-line therapy. So there's still a significant population that are non-responders. So what is non-response to standard of care? Non-response can look like an incomplete response or it can look like treatment failure. An incomplete response is an improvement in your laboratory values and histologic findings, but insufficient to truly meet the criteria for biochemical remission. So persistent elevation in your ALT. The definition of treatment failure is worsening of laboratory histologic findings despite compliance with standard of care therapy. Again, this is reported in up to 50% of patients in some studies, and there is an increased incidence of cirrhosis and the need for transplantation. In the group that meets the definition for treatment failure, mortality approaches nearly 30%. There are several factors that are associated with non-response to therapy. Age less than 40, a slower decline in your ALT, a ferritin that's less than two times upper limit of normal, IgG that's greater than 1.9 times upper limit of normal, and vitamin D deficiency have all been associated with non-response to standard of care. A recent paper by PAPE and CGH showed that a greater than 80% decline in the level of AST after eight weeks of therapy was associated with normalization or persistent normalization of liver enzymes at weeks 26 and 52. So we do think that the rapidity of decline in your serum immunotransferase has a prognostic value. So we'll discuss a case to illustrate a real scenario of which both intolerance and incomplete response were factors in a patient. So this is a patient that's 68 years old with no past medical history who presented to her PCP for fatigue and low-grade fevers for two weeks. Her liver chemistries were notable for a bilirubin of 10 AST and ALT that were markedly elevated in the 2000s in an INR of 1.6. Her laboratory workup demonstrated an elevated ANA, smooth muscle antibody, an elevated IgG, and an elevated ferritin. Her liver biopsy showed severe active interface hepatitis with grade three inflammation and stage three fibrosis, which was compatible with the diagnosis of autoimmune hepatitis. If we go back to the slide that Dr. Bitterman showed of the revised guidelines from this year, we see that she falls in the definition of acute severe autoimmune hepatitis given her coagulopathy and her cholestasis. So in this patient, remember, we don't use budesonide and we don't use azathioprine upfront. She was started on prednisone, 60 milligrams a day. With prednisone, she did have a decline in her bilirubin and her sero-immunotransferase levels, and azathioprine was added when the bilirubin had dropped to 3.8. And we normally prefer to add azathioprine when the cholestasis is improving or resolving. Unfortunately, she developed melt sores with azathioprine and had to be transitioned to MMF. So she would fall under the category of having an intolerance to azathioprine. And we would assume that with this intolerance, and I'll show you the data for this shortly, with an intolerance to azathioprine, we would hope that switching to another class of antiproliferative agent would be equally effective in terms of standard of care or first-line therapy. While her prednisone was being tapered, she had an increase in her liver enzymes. And unfortunately, MMF and prednisone alone were unable to achieve biochemical remission, which was normalization of her ALT value, and tecrolimus had to be added as a second-line agent. So just to briefly show you a graphical depiction of what happened to her, as her prednisone was decreased, Celsept was added on as a second-line agent due to intolerance to therapy. Unfortunately, while the Celsept was added on, she had a decline, but that decline sort of plateaued around this area. Now you can sort of argue that this would be, and by the old definition, this would satisfy the criteria of biochemical remission, and you don't need to do more. But we know now that this patient is still likely to progress and develop more chronic liver disease. So tecrolimus was added as a second-line agent to her. We'll review some data that looks at MMF as second-line therapy. So in this meta-analysis of 12 studies that included 397 adult patients, all of these studies were retrospective cohort studies, and the median follow-up was about 34 months. The pooled biochemical remission rate was about 58%. And five of these studies looked at the indication for why MMF was started. MMF was started for azathioprine intolerance for in about half of the cohort, and biochemical response was noted in 82%, compared to 32% that had a biochemical response when the indication was due to a non-response to standard of care therapy. So the rates of biochemical remission we think are higher if you have an azathioprine intolerance, and a little bit lower or significantly lower if the indication to starting MMF is a non-response to standard of care therapy. The pooled adverse event rate was about 14% in this population, but the discontinuation rate fell to only 8%. You can see here that most of these studies sort of were consistent in terms of the biochemical response rate. There's just one study that is a little bit outside what the biochemical remission rate. And this study had a very small number of patients and included patients that were very difficult to control in terms of the AIH. So likely a little bit of a heterogeneous population up here. In another study of a meta-analysis of 15 studies that looked at 283 patients, this showed that MMF was the most common second-line agent used. So there are options for second-line therapy that we'll review that include calcineurin inhibitors, but MMF tends to be the most popular one used. A reduction in aminotransferases occurred in about 79% of patients, so in 239 of the 283 patients. And the ranges of aminotransferase reductions varied in the studies from anywhere from 33 to 100% of patients responding. There was a significant amount of heterogeneity in terms of the definition of remission in these studies. And this is the reason why they couldn't necessarily pool the outcome for one definition as biochemical remission. Some studies used remission defined as less than two times upper limit of normal and others less than one times upper limit of normal. Circulating IgG levels were also reported in four studies in this meta-analysis. And among these studies, 50 to 100% of patients showed an improvement in their IgG levels during treatment. Two studies looked at histologic remission in patients that were started on MMF, and histologic remission was observed in 89% of patients. The adverse effects of MMF were already reviewed by Dr. Bitterman, and in these studies, the pooled reasons for adverse events leading to discontinuation included leukopenia, gastrointestinal symptoms, and headache. In another retrospective cohort study of 105 AAH patients that were managed at 17 Australian liver centers, 60% of patients achieved biochemical remission with MMF by 12 weeks. And about 60% of the cohort was started on MMF because of azathioprine intolerance, and the rates of biochemical remission were higher at 63% compared to 38% that started MMF due to a non-response. So similar to the earlier meta-analysis that we saw in terms of differences in intolerance to azathioprine and non-responses being indications to start MMF. In this table, you can see that in the patients who started MMF for non-response to standard of therapy, which is this section right here, only older age predicted a good response to second-line therapy. In those that were intolerant to standard of care therapy, a lower baseline IgG level, a lower INR, and older age at commencing MMF were associated with a complete treatment response. The bottom line from these meta-analyses and retrospective cohort studies is that MMF is probably more effective in those that are azathioprine intolerant compared to those that are non-responders to azathioprine. Now let's go back to our case quickly. As you can see here, after several months of both MMF and prednisone, the aminotransferase levels plateaued again. So our options could have been to start a second-line therapy. She was already on MMF up here. So we can talk about different calcinorin inhibitors that could have been started in her. We'll review the data for both tacrolimus and cyclosporine next. So one of the largest published experiences with tacrolimus comes from a retrospective study across 19 centers that included 201 patients in Europe, the United States, Canada, and China. In this study, MMF and tacrolimus were used as second-line therapy, and complete response was defined as biochemical remission using the stringent criteria within six months of initiation of your second-line therapy. So MMF and tacrolimus, as you can see here, maintained a biochemical remission, biochemical response rate of 92 and 94% respectively among those that were intolerant to first-line therapy. So relatively equal response rates, biochemical response rates with both MMF and tacrolimus in the group that's intolerant. However, if you look at those that were non-responders, significantly more non-responders that were treated with tacrolimus compared to MMF had a complete response. 57% of those treated with tacrolimus had a complete response compared to only 34% that were treated with MMF, which logically makes sense as you've tried antiproliferative agent already in terms of your first-line treatment and you haven't been responsive, so you likely need to use a different mechanism of action for your second-line therapy in those that are non-responders. Overall, though, the rates of biochemical remission are significantly lower in non-responders in both cohorts, MMF or tacrolimus, compared to those that are intolerant. During maintenance phase of therapy in this study, the steroids were actually withdrawn in about 26% of patients that were treated with MMF and in 20% of patients with tacrolimus. So a very important conclusion from this study is that tacrolimus may be superior to MMF as second-line therapy in AIH patients that are non-responders to standard of care therapy. In another review of seven studies with 162 adult patients that were started on tacrolimus, 121 were able to achieve normalization of aminotransferase levels. Now, the caveat to this study is that a normal ALT was defined as less than 50, so not as stringent of a cutoff. The tacrolimus doses ranged anywhere from one to six milligrams a day, as you can see in the table, and the troughs had really wide ranges, anywhere from 0.5 to 10 nanograms per milliliter. The adverse effects of tacrolimus were reported in about a quarter of patients in this study, and these included neurotoxicities, such as headache and tremor, nausea, hair loss, nephrotoxicity, hypertension, and metabolic syndrome. Let's review the other calcinone inhibitor now, so cyclosporine. Although cyclosporine has been around longer, we sort of have the same problem that Dr. Bidiman was mentioning for first-line therapy. The data for the treatment in AIH is still quite limited, and most of the data comes from small case series that's from the late 1980s to the 1990s in adults. There is slightly more data in the pediatric population, and we think that's likely because there's more frequent contraindications to using steroids in the pediatric population. So some of the data for cyclosporine, there's a compilation of 10 studies since 1985, and that included 133 patients that use cyclosporine as second-line therapy. A positive response, which just means an improvement in your immunotransferases, was reported in 93%. Again, it's hard to know what that means without being able to pool a biochemical response rate in these studies, as there's a very broad definition for what a positive response meant. An important thing to note is that cyclosporine can be dosed at very subtoxic trough levels to likely achieve a positive response in these studies. In pediatric study of children treated with cyclosporine as second-line therapy, eight were treated out of 21 with second-line therapy to achieve biochemical remission, and trough levels were initially about 150 to 200 to induce remission, and then were gradually able to fall to around 50 after a year. This broad range of trough levels is considered an important benefit of cyclosporine compared to tacrolimus, mainly because you can allow for finer dose adjustments to cyclosporine and treat it at subtoxic trough levels. And at the narrow ranges that we see for tacrolimus, we normally can't check levels less than one in our labs. They really limit the ability to truly reduce the tacrolimus to the lowest possible effective doses and then minimizing some of the toxicities. And unfortunately, calcinone inhibitors overall have yet to be rigorously studied in large clinical trials for AIH. And so the dosing, the monitoring, the safety profiles in our AIH population haven't been fully established despite the longstanding use in this disease. When discussing with the AIH patient in front of you which second-line agent to start, it's important to individualize this discussion and the decision, and to consider the side effect profile of all the options that you can have for second-line therapy. So we talked about how the indication for starting second-line therapy might lead you one way or the other. So if you are intolerant to azathioprine, then you would probably consider using a antiproliferative agent next if appropriate, versus if you're a non-responder to azathioprine, then it might make more sense to consider using a calcinone inhibitor if appropriate. So while NMF has less metabolic complications compared to the calcinone inhibitors, it is teratogenic, and you really have to have a discussion with a patient in front of you regarding childbearing status and considerations for pregnancy. So that's probably one of the most important considerations to make with thinking about NMF as second-line therapy. Both calcinone inhibitors have significant metabolic complications, which is important to take into consideration with the comorbid NAFLD that we see, the non-alcoholic fatty liver disease that we see in our AIH population now. The metabolic complications, as you can see, include hypertension, hyperlipidemia, and nephrotoxicity as well. The degree of nephrotoxicity may be slightly lower with cyclosporine if you're able to maintain lower trough values as the laboratory is able to give us lower trough values, but we think with prolonged exposure, there likely isn't too much of a clinically significant difference with nephrotoxic effects from this in cyclosporine. But these are the three drugs that are the mainstay of second-line therapy at this time, and these should be your next go-to line of drugs when you're not responding to azathioprine or prednisone. You will encounter some studies that have looked at TNF-alpha inhibitors as second-line agents, and the literature and the evidence on this is very sparse for its use as second-line therapy. In the largest single-center retrospective analysis, there are 11 patients only that have been treated with infliximab. Six patients were able to achieve normalization of their immunotransferases, but seven developed infectious complications. So given the risk of infections, especially in those with cirrhosis, infliximab and TNF-alpha inhibitors are not recommended as second-line agents right now. We will end the webinar with one additional case here. So this is a 55-year-old woman with ITP who was admitted with gum bleeding. Her bilirubin at presentation was 1.4. Immunotransferases were 601,104 and INR was 1.1. Her platelet count was 13,000. Her laboratory workup demonstrated an elevated ANA, smooth muscle antibody, an IgG level, and an elevated ferritin. Her liver biopsy after receiving platelet transfusions showed severe active hepatitis with a lymphoplasmicytic inflammatory infiltrate pericentrally with bridging necrosis. She had grade four activity and stage four fibrosis. She was initially treated with prednisone, as you can see here, at a mg per kg per day and was admitted to the hospital. And with the prednisone, the green bar here is your ALT and the orange is the platelet count. With the treatment of prednisone for both her ITP and autoimmune hepatitis, her ALT declined nicely as her platelet count started to rise. And she was eventually discharged on azathioprine and a prednisone taper. As her prednisone was tapered, her azathioprine was increased to 100 milligrams per day and unfortunately, several months later, she had another flare with her platelet counts dropped and her ALT began to rise again. The second time around when she was admitted to the hospital, she didn't respond to other therapy for her ITP and was eventually started on rituximab. After she received several doses of rituximab in the hospital, she had a nice decline in her ALT value to baseline levels and her platelet count has remained normal. She was admitted to the hospital for biochemical remission three years after she received her dose of rituximab. So the largest experience of rituximab comes from the International Autoimmune Hepatitis Group. This included 22 autoimmune hepatitis patients that didn't respond to first or second line therapy. So you've failed both your first line agents, standard of care therapy, you've maybe have tried MMF or calcineurin inhibitors and these patients were then subsequently treated with rituximab. The most recent cases were observed one month after rituximab infusion and sustained for nearly two years. About 95% of patients had no flare within a year and 71% had no flare and maintained remission at two years. Two thirds of the population was also able to reduce their steroid dose. So these findings have laid the groundwork to study additional B-cell targeting therapies such as enolimab, or whether or not rituximab is an appropriate third line agent or an appropriate agent to use. But it did lay the groundwork to study other B-cell depleting therapies. To summarize the key points from this webinar, remember the goal of AIH treatment is to achieve biochemical remission, which is now defined as normalization of your immunotransferase levels and IgG with complete elimination of histologic activity. And the map should be considered an alternative first line therapy in specific situations only, including a low likelihood of patients tolerating high dose steroids, which is rare, if a patient has complete TPMT deficiencies, or if you're worried about preexisting cytopenias or hyposplenism. There isn't strong data yet to support MMF as first line standard of care therapy right now. Intolerance to first line therapy is reported in up to 20% of patients and MMF and tacrolimus have a similar efficacy in achieving biochemical remission in this cohort. Non-response to first line therapy is reported in up to 50% of patients and tacrolimus likely is more effective than MMF in achieving biochemical remission in this cohort. It is important that we individualize treatment based on the expected side effect profile of second line therapies. And we consider concurrent things like family planning and metabolic syndrome in our population. Much of our data, as you can see that we presented here is from a compilation of case series or retrospective cohort studies. And the data has been pooled and combined and sliced and diced in different ways with different analytic methods. But what we truly need is better randomized control studies to understand how to best manage this disease and to create further therapeutics. So we still have much to learn about this disease. Thank you for attending this webinar today and we'll open it up for Q&A now. So we have a question that's come from an audience member. How do we distinguish between non-compliance that the patient doesn't want to reveal versus a partial response? That's a really good question. Dr. Bitterman had mentioned some of the metabolites that we can check for and I'll let her review what she had mentioned earlier. Sometimes some of the metabolites, it's gonna be a little challenging for me to show you the pathway, but if you remember from the pathway, you can check certain metabolites to ensure that the patient is in fact taking the medication. And potentially you could imagine if your active metabolites are high, but let's say the transaminases are also uncontrolled, then potentially this would indicate a partial response versus if the metabolites are low, then potentially the patient is missing doses for instance. Another question that's coming from the audience is how long do you wait to decide to change first line to second line therapy, six months? And do you biopsy all patients that don't respond? That's a really good question. And there isn't a strict cutoff for the amount of time that you should wait before you decide that your standard of care is not working and that you really need to jump to starting second line agents. If you remember the different patterns of decline that you can see with the graphs that I showed, some patients will have a very slow steady decline that will take longer than six months. Some patients will have a very rapid decline that occurs incredibly quickly within eight weeks. And those patients likely have a better prognosis to responding to first line therapy. So there isn't a clear time cutoff. I think what matters the most is actually the trend that you're seeing. So if a patient's ALT is continuing to decline, but it is still upper limit of normal at six months, that's not a reason to add second line therapy. I would continue to watch until you really see what the plateau is and what the maximal effect of your first line therapy and standard of care therapy is. Once you reach that plateau, and if that's still elevated from upper limit and normal, then it's time to add second line agents. So there isn't a clear cutoff for do you wait six months, do you wait two years? And the literature from a lot of these cohort studies that we described in these case series is all over the place for defining, because the study has to define an endpoint. So some studies will define it at six months, some studies will define it at three years. So the literature is all over the place. I think clinically what matters the most is the pattern that you're seeing. And when you start to see that plateau, where that plateau is occurring. And similar to that, I think that there's a group of questions coming in about if a patient is non-responsive to MMF or to first line therapy, do you continue MMF and azathioprine and add on to Crolimus? Do you stop MMF and then just do to Crolimus? So usually if you are non-response because of an incomplete response to therapy, you would add on a second line agent just to allow for the dual mechanism of action of an antiproliferative and a calcineurin inhibitor. And in some patients, you might even still have a low dose corticosteroid. So it is targeting at different mechanisms of action to control the disease. You don't necessarily abandon one class unless you're having significant side effects of the medication. Maybe we can talk, and I think maybe Dr. Vierling has a lot of experience with this and can briefly answer a group of questions that's related to optimal to Crolimus trough levels and therapy. And maybe you can also describe optimal cyclosporine trough levels when you're adding it for second line therapy. There is very wide ranges in the literature. And I think that Dr. Vierling probably has the most experience with that. Well, first let me thank both of you for such an excellent lecture. You've gotten into the details and revealed not only what we know, but the urgency that we should all feel as a hepatology community to band together to answer through appropriately powered randomized controlled trials what we need to know in the future to help our patients. We lack those as you've emphasized. So to your question, when one uses a calcineurin inhibition, it was always predicated as we're all aware on the use in transplantation to prevent a solid organ transplant rejection episodes. The amount of immunosuppression required for that far exceeds in most everyone's experience that which is required to actually manage an autoimmune disease, be it autoimmune hepatitis or other diseases of autoimmunity. So in general, we should have the concept that lower doses than those used and therefore lower trough levels than those used in transplantation could be perfectly sufficient. The best data you actually went over Dr. Goel showed that suboptimal trough levels of cyclosporine were often completely satisfactory in achieving complete remission. And that owes to the fact that the trough levels of cyclosporine for transplant immediately afterwards about 250 nanograms per ml and you wanna keep it above 125, well, you can manage patients and we certainly have our own anecdotal data between 50 and 75 nanograms per ml where there's no real issue of toxicity with related to hypertension, nephrotoxicity, gingival hyperplasia, et cetera. However, with jecrolomus, we really don't know how to manage because by inference, you'd suspect that you could manage patients somewhere between one or one and a half nanograms per ml. Our assays are being devoted to transplantation community needs really don't even report levels less than two in our current assay in our transplant program. So we don't have the fine tuning ability and the therapeutic ratio of the effect versus toxicity is wanting in jecrolomus. Lastly, I think it's imperative that whenever a person uses a calcine or an inhibitor, that you individualize it to your patient. And that the best bottom line certainly for jecrolomus is that if you are changing by 0.3, your creatinine, you are taking too much jecrolomus. We can't exchange a progressive risk of chronic kidney disease in patients who are being treated with autoimmune hepatitis, usually in mid adulthood or later adulthood. So creatinine is a bit of a biomarker for jecrolomus and cyclosporine dose relations. Thank you, Dr. Bierling. I love the historical context as well. I appreciate it. There's a group of questions that are coming in related to metabolism of azathioprine and adequate dosing. And if we target appropriate 6TG levels in our population, and if we use allopurinol to shunt, if we ever have lower levels. Dr. Bitterman, maybe you can talk a little bit about the use of allopurinol and just review the mechanism with the xanthine oxidase inhibitor. Yeah, so I have not used that in practice myself, but I have heard of physicians using allopurinol. If you recall, it's sort of a similar effect to TPMT deficiency. Essentially, what happens is that by providing, by giving a xanthine oxidase inhibitor, you have increased production of the active metabolite from azathioprine at a given dose. And we do know that there are some patients who are sort of rapid metabolizers of azathioprine. And so who may actually be generating lower doses, lower levels of 6TGN at a sort of comparable dose. And so potentially the use of allopurinol could be used. I don't know if Dr. Veerling, if you have any comments on that in particular, that aspect. Well, yes, I think you've covered the mechanism. So by blocking xanthine oxidase, you're going to redirect metabolism, not to 6-thiouridine, but onward downstream to 6-MMF-MFP and thiobuonine nucleotides, which of course are the immunosuppressive nucleotides. We're not where the IBD community is in trying to target actual TGN levels for therapy. We've never really had those measurements, to my knowledge, in autoimmune hepatitis. But the largest series advocating allopurinol use for those that have excessive 6-thiouridine metabolites comes from Intel de Boyer, who published a series in which he showed that the use of 100 milligrams of allopurinol daily redirected to give substantial increases in all but one, if I recall it correctly, a patient who had low thiobuonine nucleotide immunosuppressive levels, redirected the 6-thiouridine to that pathway and biochemically improved them all. So there really is some data, anecdotally again in a series, to suggest that this would be a way to avoid abandonment of azathioprine because of a perceived lack of full responsiveness. I think the only thing I'll add is that it's important that if you are going to start allopurinol to shunt your metabolism a little bit more towards the active metabolite is to reduce the dose of azathioprine that you're giving because you can actually create some very severe cytopenias in that population then. So if you are going to start allopurinol to decrease the 6-MMP metabolite and to increase the 6-TGN, reduce the azathioprine dose by 50%. That's what we do in our IBD population as well just to prevent the cytopenias. You might need to increase it again, but just to start off, make sure you reduce the dose. Right. There's another group of questions that are coming in related to first-line therapy actually. So are we okay if azathioprine is maintaining, if azathioprine at a certain dose is maintaining remission with a low dose of prednisone, are we okay continuing that low dose of prednisone or should we add second-line therapy? Is that an okay goal? And I think that's a very relevant question. We do obviously see that a lot clinically. And I'm very curious to hear about Dr. Bitterman's and Dr. Veerling's suggestions for this. The one thing that I would consider is just what the comorbid illnesses are in terms of the tolerance to a second-line agent. So if this is a patient that already has fatty liver disease or some kidney disease, then I'd be a little bit nervous about it. You wouldn't add MMF on top of azathioprine and prednisone. So you would be thinking about adding a calcinorin inhibitor. The other option could be if you're just on a low dose of prednisone is could you get away with low dose of budesonide in that population to avoid some of the systemic toxicity with the steroid. But I'm curious to hear about Dr. Bitterman's and Dr. Veerling's opinions. Yeah, and I think actually this relates to another question in the chat about any preference in management when there's co-existent NAFLDs. So this is actually an area of research interest of mine. We see more and more patients who they just happen to have metabolic risk factors because we know these are more prevalent in the population, including fatty liver disease. And they may even have, I have that first case I presented actually is a patient of mine. She had steatosis and steatohepatitis on her initial biopsy. And the reason that becomes relevant also is that when you think about normalization of enzymes, it might be hard to distinguish whether the failure to normalize enzymes is truly AIH disease activity or potentially steatohepatitis. And so I think, you know, ultimately the goal increasingly is this desire to try to minimize steroids as best as one can. And I agree with sort of the potential for budesonide if that is effective and the patient does not have cirrhosis. But I think it's as, you know, as Dr. Coyle mentioned, it's very much a, you have to take an individualized approach and, you know, patients are complex and they have lots of already preexisting comorbidities that may, for instance, may make steroids more desirable and a calcineurin inhibitor less desirable. So it really, I think it has to be individualized. I think those are excellent points. Our writing group determined and published in the 2019 ASLD AIH guidelines, an overall composite estimate of 30% of our adult AIH patients have NAFLD in the United States. And that fits obviously with what we say about the adults in the United States in general having that risk. One of the most recent studies, although it's small, it was at least three centers, hepatology biased referral centers. They found that 30% of their patients had NAFLD and essentially half of them had NASH on biopsy. And so that becomes sobering because we have something that we would expect can raise the transaminases independently of how we may manage AIH inflammatory processes generating aminotransferase elevations, just as you said. Now, with that group, I think that stepping back on first principles, if someone has achieved a deep and sustained remission on low dose prednisone and azathioprine or for that matter, 6-MP, one could further reduce the prednisone with an idea that maybe monotherapy with the antiproliferative would be sufficient to maintain remission. So that's one option. The other option, if they need low dose steroids, the non-steroidic would be buprenesinide. But then if it's looking like you need both and that you're still not perhaps achieving the deep remission of our guidance, then considering a second life therapy like a calcineurin inhibitor comes into play. There, we really have to be very careful. If we've got patients that do have a phenotype of NAPL or NASH, or known to have NASH on biopsy criteria, then we have to keep in mind the diabetic genetic potential of decrolimus as opposed to cyclosporine. That might be a point of differentiation for individualization of the choice of the calcineurin inhibitor. Although in contrast cyclosporine will have more effects on lipids, those can be managed by other strategies. But to add injury to insult to type two diabetes risk in the metabolic syndrome with a diabetic drug, decrolimus might be an error. Thank you both. I think sort of related to the comorbid illnesses, with NAFLD is the question about whether or not concurrent fatty liver disease reduces the progression of autoimmune hepatitis and reduces the response to therapy. I actually don't know if we have data to really support that. I think usually we think about concurrent illnesses, obviously accelerating the rate of progression of liver disease, but I don't know if we have data about altering response to therapy. Dr. Bitterman, do you have any? I agree on the point about therapy. There is one study from a couple of years ago now that looked at coexisting NASH and EIH and did find a more rapid, I believe it was more rapid progression to cirrhosis and also potentially to the need for transplantation. I don't believe the study tackled issues related to treatment. However, it was purely just sort of looking at the natural history. I think that we probably have time for a few more questions. A question from my colleague that I think Dr. Bitterman, maybe you can try to answer in terms of metabolism. If someone is responding to azathioprine and has good 6TG levels, but you detect that 6MMP levels are high on your labs, do you need to alter or adjust the dose at all? And I'm assuming here that the liver enzymes are actually fine despite the 6MMP levels being high. So do you alter therapy just based on what the metabolites are? I'm going to defer to Dr. Vierling. Dr. Vierling, what would you do? Well, you know, there's an old principle. You might want to let sleeping dogs lie if you've achieved your biochemical remission. And I would say first that those that achieve biochemical remission, we don't have any large-scale data that is prospectively analyzed. They're thiopurine-derived metabolites. So we just don't know. I think that you can expect metabolites of a certain concentration in all the classes because it's a multi-step process of metabolism and it all doesn't shunt directly or exclusively to 6-thioguanine nucleotides as an immunosuppressant. So if your level of 6-thioguanine nucleotides appears good, your response is good. I don't think I would be independently worried by the other metabolites unless there was a suspicion of some metabolite-related toxicity. In that case, then we're back to what I'd mentioned before. We do not know, in contrast to our IBD colleagues, what targeted range, what are the 95% confidence intervals around which we have comfort that we've generated adequate 6-thioguanine nucleotides as an immunosuppressive contribution of azathioprine. You could alter your doses in the case of the toxicity and play it empirically, but I don't think there's any guidance as to how to do so. And does dividing doses of the azathioprine at all affect the metabolite levels? Do we know that? No, I'm pondering that. I don't know the answer to that. But always remember that once you start to divide doses of anything, then we are on a bit of an issue regarding adherence, and not the intent to adhere, but just the difficulty of adhering when you've left someplace and left your drug and didn't take it on time and don't want to double up later, et cetera. We all know the issues. So there is a dose in the morning of your immunosuppressants has a contribution to adherence. Great, thank you. The last group of questions that we'll answer, there was a few questions on Rituximab, and I do just want to, I included the Rituximab data in our presentation because you will encounter it when you're searching for second, third line therapies to use or just rescue therapies to use for your autoimmune hepatitis patients. But I do want to reiterate that it is not prime time and it's not definitely ready to use for our patients with autoimmune hepatitis. We really don't have the data to support its use. In the studies that I mentioned in the larger study from the International Autoimmune Hepatitis Group, where 22 patients were treated with Rituximab, so 71% of those patients that had a flare at three years, a significant proportion of them were re-dosed with Rituximab and in a proportion, again, were able to get back into biochemical remission. So when they did flare, they were re-dosed. And what the maintenance agent is after you're treated on Rituximab is very unclear, and we don't really know if you even need anything. Some people will continue a low-dose azathioprine. That's what happened for our patient that I mentioned in the webinar, but we really just don't have any data to guide us one way or the other on what we need to do for our patients that are treated with Rituximab and if they need additional maintenance therapy and who does respond well and who doesn't respond well. I think the reason it was nice in our case is that she had another indication to receive Rituximab, which was the ITP. So we sort of saw the natural history of what happened, or the history of what happened to her autoimmune hepatitis in response to the Rituximab. That's about 20 minutes of questions. Dr. Vierling, is there any last minute point that you feel like we should make to the group? Let me just add to the Rituximab story. I think anti-B cell therapy has captivated people's attention because it was a fusion with a very long afterlife of remission and improved biochemistries. I just want everybody to know that internationally, there's now enrolling the first ever second line therapy randomized control trial, which is appropriately designed and powered and to look at a strategy of anti-B cell therapy based on monoclonal antibody against the B cell activating factor, which is elevated in the circulation of patients with AIH, as well as virtually all known autoimmune diseases. So this not only scavenges this factor, which is signaling B cells, but it interferes with its receptor and its signaling and its generation of receptors in the future, and may be more of a long vibe. If patients have failed therapy, partial responses, non-responses, instead of diving into empiric second line therapy, at least consider talking to them about their interest in learning more about this study of anti-BAPA monoclonal antibody therapy, a placebo controlled randomized trial. Thank you so much for the insight, Dr. Bierling and Dr. Bitterman for participating with us. We hope you all learned something. Have a great day, everybody. Okay, all the best.

second-line therapies

MMF

tacrolimus

AIH

biochemical response

non-responders

complete response

steroid withdrawal

dosing guidelines

individualized treatment plans