Hello, everyone, and welcome to the 2021 LIBER Meeting ASLD NASPAGAN Symposium. I'm Mercedes Martinez, a professor of pediatric and medicine at Columbia University in New York and the chair of the NASPAGAN Hepatology Committee. I co-chair this session with Dr. Amber Hildreth, a pediatric transplant hepatologist at University of California, San Diego, and a member of the NASPAGAN Hepatology Committee. At this exciting symposium, we will explore innovative management strategies for pediatric patients with autoimmune liver diseases. A world-renowned lineup of faculty will review the pathophysiology of pediatric autoimmune hepatitis and primary sclerosing cholangitis, the diagnosis, management, and risk stratification. These are six talks that will focus on diagnosis, primary and secondary line of therapy, and complex cases of autoimmune hepatitis. In addition, the speakers will review the diagnosis and pediatric-specific predictor models of primary sclerosing cholangitis and the impact of patient and allograft survival on post-transplant disease recurrence. The faculty will introduce themselves and their topics at the start of their presentation. These presentations are recorded and will be available on the Liver Meeting Digital Experience on Demand for up to 90 days after the meeting. As you view the presentations, we encourage that you post questions for the presenters in the chat box to the right of your screen. Presenters are here and ready to answer your questions in real time using the online chat feature during the stream presentation. I hope you enjoy the presentations and see you at the end of the discussion. Enjoy these outstanding speakers. I would like to thank the organizers of the ASAD and Aspen Pediatric Symposium for inviting me to talk with you about the impact of genetic predisposition and environmental triggers in the pathophysiology of autoimmune liver disease. That's truly a topic near and dear to my heart. These are my disclosures that are relevant for this talk. It's true for many other health problems, autoimmune liver disease also arises from a complex interplay of genetic predisposition and environmental exposure. Technical advances in next-generation and pyrosequencing and development of multi-center collaborations to study larger numbers of patients with autoimmune hepatitis or PLC have helped us to uncover some exciting insights into etiologic factors of autoimmune liver disease. In the first part of my talk, I will review recent publications on certain medications linked to autoimmune hepatitis, followed by a very recent publication on the intestinal microbiome in patients with PLC and how it may shape disease progressions. In the second part of the talk, I will review some of the results of our whole exome sequencing studies in patients with autoimmune liver disease in Cincinnati, and I will discuss two genes in particular, the autoimmune regulator gene and TNF-ARP3. Cara Mack and others recently published an updated guideline for the management of autoimmune hepatitis, and that document includes a nice table which shows several medications which have either a definite, probable, or possible association with autoimmune liver disease. Many of you, of course, have seen in your own practice patients who developed autoimmune hepatitis following the exposure to minocycline. Infliximab is an emerging problem, and one particular problem is whether we should treat these patients with steroids or not. The investigators from the National University Hospital of Iceland recently published their results of reviewing a database which allows them to study all patients in Iceland treated with infliximab, and they found 36 patients who developed liver injury following the exposure, and they compared patients who were either treated with corticosteroids, 17 of them, or not treated with corticosteroids. They found that essentially both groups had a similar total duration of the disease, but patients who were treated with corticosteroids more quickly resolved their elevated LFTs following peak injury. Here is a table which I prepared one and a half years ago for the emerging topic conference on nuclear receptors and the role of the microbiome in cholestatic liver diseases, and that turned out to be one of the very first ASAD functions which was canceled because of the pandemic. It was very sad because it was an excellent program. At least now I can come back to that table because it turns out that one of these studies, the studies by the group from Norway and Germany which showed findings consistent with other previous studies that there was a decreased alpha diversity of the intestinal microbiome in patients with PSC and an abundance of a particular species, Valenella, that study was taken further by investigators, and they went back to those fecal DNA samples and performed metagenomic shotgun sequencing, and that metagenomic shotgun sequencing allowed them insights into the microbial function, and they found that patients with PSC showed a marked difference in the abundance of genes related to vitamin B6 synthesis and branched-chain amino acid synthesis. Then they went to plasma samples from an independent cohort and performed targeted metabolomics on these plasma samples, and they did find that patients with PSC had lower concentration of vitamin B6 in the plasma or certain branched-chain amino acids. More importantly, when they looked at the particular levels and they found a correlation between the levels in transplant-free survivors, as you can see in the top right, patients with plasma vitamin B6 levels in the lower two quarters had a much reduced transplant-free survivor compared with those within the upper two quarters. What do we know about the genetics of autoimmune liver disease? Most of our knowledge originates from genome-wide association study in larger cohorts of adult patients with PSC or autoimmune hepatitis, and I show here the Manhattan plot to summarize all the susceptibility genes in various particular immune genes related to PSC and the strong HLA predisposition autoimmune hepatitis. Of course, in pediatrics, we always hypothesize that the early onset may be related to monogenetic causes, putting particular genes on the map which may have a strong effect on disease pathogenesis, and that was a hypothesis which informs the rationale for us to do such whole exome sequencing in our patients with autoimmune hepatitis and PSC. We did it in 42. The variant calling was performed by a collaborator, Alex Valencia, who focused on patients and genes with variants which were rare and healthy controls, less than 1%, which affected the amino acid sequence and which were predicted to be damaging. And then we focused on patients which either had two variants in genes associated with autosomal recessive disorder or one variant in genes associated with autosomal dominant disease. And here, I show you an overview of our results. Forty-eight patients were enrolled into the study. Six patients had variants detected by clinical testing which were found to be relevant, and whole exome or whole genome sequencing was performed in 42 patients. In the top part, I show you the results on three patients who carried rare biallelic variants in genes potentially associated with liver disease or autoimmune liver disease in AMBERS, DOCK8, and AIR, and I will talk about AIR a little bit more in detail afterwards. What was unexpected is that we found rare heterozygous variants in a fair number of patients, and these genes were linked to autoinflammatory condition and interferonopathy. Rare variants in TNF-ARP3 were found in four patients and PSG2 in four patients, not two in two patients. We also found one patient with a rare variant in GATA2, which controls T lymphocyte polarization, and that, in particular, gene has been linked to autoimmune hepatitis by the group from Birmingham before. We found genes…variants in genes associated with Kabuki syndrome, which control complement function in one patient with a NODGE2 mutation, which is, of course, related to bile duct development. Here's our patients who were subsequently found to have a mutation in AIR. That patient presented with a history of hyperparasitism, hypothyroidism, pancreatic exocrine deficiency, candidiasophagitis, so very typical for autoimmune polyendocrine…polyendocrinopathy syndrome, and she presented to my clinic with an acute rise in AST, ALT, alkaline phosphatase, and GGT. Her IgG levels were highly elevated. Her autoantibodies were surprisingly negative. The liver biopsy showed interface hepatitis, advanced inflammation, plasma cells, and early fibrosis. There was some pericolangitis. We treated the patients with budesonide, azacybin, and ursodile, and the LFTs quickly improved. That patient was found to carry a homozygous deleterious mutation in the autoimmune regulator gene, and that is, of course, very famous and well-known to control the negative selection of autoreactive T cells in the thymus. If that is defective, then those autoreactive T cells tend to cause a variety of problems, in particular, immune-mediated destruction of the endocrine organs and liver disease. There was recently a publication by the NIH on their patients with APS1 and mutations in AIR. They found, in this cohort of 43 patients, autoimmune hepatitis in about half of the patients. Interestingly, the majority of these patients did not have any autoantibodies, and that is striking to me, given how much stake we put in the diagnostic relevance of autoantibodies for autoimmune hepatitis. The particular mutation we found in our patient is very common patients who develop autoimmune hepatitis in the context of APS, and in that cohort, patients typically, although it's such a strong underlying genetic predisposition, that patients usually responded quite well to immunovigilator therapy with azacybin or corticosteroids. Now, I would like to shift to another gene, which we found in four of our patients, TNF-AAP3, which encodes A20, a zinc finger protein, which is rapidly induced by TNF-alpha and inhibits NF-kappa-B activation and TNF-alpha-mediated apoptosis. Heterozygous deleterious mutations are associated with autosomal demonant Beschetz disease and very early onset IBD. These pictures show some of the manifestations in patients with this rheumatologic disorder of Beschetz disease. So what are our patients with liver disease? We have four patients. Patient number one presented with type 2 autoimmune hepatitis, strong LKM-positive antibodies, and a presentation more of an acute liver failure at early age. I show some of the findings of the histology and some of our immune workup. The other patient presented at age 10 with seronegative autoimmune hepatitis and was subsequently admitted several times for multisystem inflammatory syndrome presentation with rash, fevers. That patient also developed Hashimoto's thyroiditis. On the table on the right, you can see four patients who belong to one family. The older two siblings, the oldest son presented at age 13 with type 2 autoimmune hepatitis with LC1-positive antibodies, already with cirrhosis at presentation. His sister presented at a similar age with a milder type of autoimmune hepatitis, quite responsive to therapy, but then she developed those Beschetz-like disease with oral and genital ulcers and joint manifestation. One brother carries the same mutation in TNF-LAP3, which is a heterozygous deletetion of the entire coding sequence. That patient has normal LFTs, but recurrent aftersouces, and the mother who carries also that mutation and inherited it to her children has a waxing and waning course of elevated LFTs and various forms of immune dysregulation. Here's some images of the liver biopsies of the four patients, and you can see strong inflammation and a variable degree of fibrosis, which was, of course, advanced in one of the patients at presentation. In summary, TNF-LAP3 mutations confer risk to type 2 autoimmune hepatitis or seronegative autoimmune hepatitis multisystem inflammatory disorder. You may remember that several patients had variants and genes associated with interferonopathy. Rheumatologists care about these diseases, and they recently developed a 28 type 1 interferon response gene panel, which helps to make the diagnosis and select particular therapies, and we looked at the gene expression for these interferon associated genes in our patients. This is a very early findings, but you can see that patients' healthy control segregated with low expressions, whereas at least some of the patients, those with PRCG2 mutation and those with IFIH1 mutation had a very similar pattern of gene expression up or down regulation in this particular panel. So, the summary of surveillance for monogenetic causes for autoimmune liver disease shows that we had a few patients with biallelic mutations, heterozygous variants were found in a surprising 58% of our patients. Of course, we need much more work to truly link those genes with autoimmune hepatitis or PSC. We have strong evidence for AIR-GADA2 and TNF-ARP3. We have early evidence that interferon signature may be helpful in the future. And now here, I would like to conclude with making you aware of some research opportunities which will soon happen in 12 of our centers in the US and Canada. We will start enrolling patients with PSC into a prospective study and that will give ample opportunity to validate those findings from adults on FECA pyrosequencing, on plasma metabolomics, and of course, we will have DNA for whole genome sequencing, whole exome sequencing. We will carefully phenotype these patients with CL, MRCPs, MR elastography. And I hope we will learn over the next years how pediatric liver disease is driven by genetic and environmental factors. In summary, environmental triggers for autoimmune liver disease continue to emerge and require us to be vigilant. Our understanding of the gut-liver axis in autoimmune liver disease has progressed from characterizing abundance of microbial communities to linking their metabolism to circulating nutrient metabolites and monogenetic conditions related to control of inflammation might predispose to pediatric onset autoimmune hepatitis. I thank you for your attention and I'm ready for questions. Hi, I'd like to thank Dr. Martinez and the ASLD for inviting me to speak on approach and treatment of typical cases of autoimmune hepatitis. Here's my disclosure slide. The objectives of this talk will be to go over treatment options based on the mode of presentation of autoimmune hepatitis. This includes typical cases, acute severe autoimmune hepatitis, the cirrhosis presentation of autoimmune hepatitis, and drug-induced autoimmune hepatitis-like injury. I will not be discussing acute liver failure presentation as Dr. Martinez will highlight that in her talk. Next I'll go over other considerations regarding the initial management and many of these come from the 2019 ASLD guidance statements. And finally, I'll finish with information related to treatment withdrawal. The diagnosis of autoimmune hepatitis entails the characteristic histologic abnormalities such as lymphoplasmicytic interface hepatitis as shown on the right. The diagnosis includes elevated AST and ALT and elevated total IgG and or the presence of one or more autoantibodies. Shown here are the various treatment response definitions associated with autoimmune hepatitis. Biochemical remission is normalization of AST, ALT, and IgG. Neurotoxic remission includes the absence of liver tissue inflammation while on treatment. Now the other response definitions from treatment failure through treatment intolerance I will not be discussing as Dr. Mohamed will highlight these scenarios in his talk. The treatment of typical autoimmune hepatitis includes induction therapy with steroids, either prednisone or budesonide, and maintenance therapy with azathioprine. It's important to note that it is recommended that for azathioprine the TPMT activity be measured prior to initiating azathioprine based on the fact that 0.5% of individuals will have absent TPMT activity and are at high risk for severe bone marrow suppression. After one to two weeks of induction therapy with the steroids, it is then recommended to initiate the azathioprine. A word about budesonide as an alternative to prednisone for induction therapy. Based on the many systemic side effects of prednisone, it is recommended to definitely consider budesonide in the following scenarios. The patient with obesity, fatty liver disease, and importantly with fatty liver disease, it's been recently reported that up to 30% of autoimmune hepatitis patients will also have fatty liver disease. And if you have both diseases together, you have significant increased risk of liver related mortality and adverse outcomes. Budesonide should also be considered for your patients with significant mental health disease and severe acne. Moving on to acute severe autoimmune hepatitis presentation. The definition of acute severe autoimmune hepatitis includes jaundice, an INR of greater than or equal to 1.5 but less than 2, and no encephalopathy. It's important to have close surveillance for acute severe autoimmune hepatitis due to the increased risk of progression to acute liver failure. On the right is the histology of acute severe presentation, and it has unique characteristic features distinct from your typical autoimmune hepatitis. These include central perivenulitis, central lobular hemorrhagic necrosis, and the lack of significant fibrosis. The treatment of acute severe autoimmune hepatitis includes prednisone for induction. Budesonide is not recommended, and this is due to the fact that there have not been any studies on the efficacy of budesonide for acute severe presentation. Azathioprine can be started after the first week or so. And as I mentioned, you have to have close monitoring in the acute severe presentation, including considering daily liver tests in the first 7 to 14 days due to the high risk of evolving to acute liver failure. Moving on to the treatment of autoimmune hepatitis with cirrhosis, again prednisone is recommended for induction, and similar to acute severe presentation, budesonide is not recommended. And the reason for this is twofold. First, in the setting of cirrhosis with portal hypertension, the portosystemic shunts will decrease drug efficacy, and due to the fact that it's not being properly metabolized, budesonide could have increased steroid side effects. Secondly, there have been reports that budesonide and cirrhosis may increase your risk of portal vein thrombosis. Azathioprine is not recommended in decompensated cirrhosis, but can be used in compensated cirrhosis. One thing to remember there is that there has been a reported increased risk of azathioprine-induced cytopenias with cirrhosis. In general, laboratory monitoring after you initiate steroids will include checking the liver tests every one to two weeks. Once your AST and ALT have significantly improved and are coming down towards normal, then you can start your wean of steroids. And again, you're going to be wanting to check laboratories every two weeks, and once you're in biochemical remission, you can space it out to eventually quarterly. The goal is to be steroid-free in biochemical remission by six months after the diagnosis and on monotherapy thereafter with azathioprine. I wanted to say a couple words about drug-induced autoimmune hepatitis-like injury. Shown here is the list of definite, probable, or possible drug associations with the development of autoimmune hepatitis-like disease, and I've circled the most common drugs in pediatrics that we see, and by far, minocycline-induced injury is most common in pediatrics. The treatment for drug-induced autoimmune injury includes withdrawal of the offending agent, and if it's a mild injury, you can observe for normalization of liver tests, which usually happens within a month of stopping the drug. When do we consider steroid therapy? We consider steroids if the injury meets criteria for High's Law with an AST or ALT of greater than three times upper limited normal and a bilirubin of greater than two times upper limited normal. We'll also consider steroid therapy if there's no improvement or worsening of those liver tests after you stop the offending agent. Now, if you are on steroid therapy and you wean the patient off, if the liver tests rise again, then it's recommended that you manage that patient like a typical autoimmune hepatitis. I would be remiss if I didn't educate everyone on the recent reports of COVID vaccine-induced autoimmune hepatitis. To date, there have been seven cases reported. It is usually a very acute onset of liver symptoms within a week of the vaccine. The liver histology is consistent with autoimmune hepatitis, and autoantibodies are present, and the treatment includes steroids with or without azathioprine. On the right is an example of a patient who had COVID-induced autoimmune hepatitis. As you can see, after receiving both vaccine doses, there was a significant rise in liver tests and the liver abnormalities resolved with the initiation of prednisone. Moving on to additional considerations for the management. First, importantly, there should be a high index of suspicion for concurrent autoimmune diseases, which can occur in up to 40% of patients with autoimmune hepatitis. The two most common are celiac disease and thyroid disease, and therefore, it's recommended to screen for these at the time of diagnosis. Furthermore, it's recommended to screen for many different autoimmune diseases, such as IBD or diabetes, based on symptomatology. Finally, one should consider working up for overlap with PSC. Some institutions do this up front and get an MRCP at the time of diagnosis of autoimmune hepatitis, and others will consider it if the patient does not respond to standard therapy. Other considerations, including supplementing with calcium and vitamin D while on steroids. Second, it's recommended that you defer your transient elastography for six months after diagnosis based on the fact that the inflammation can also increase liver stiffness and potentially overrepresent the degree of fibrosis that's actually there. And overall, you also want to And overall, you also want to ensure that your patients are vaccinated for hepatitis A, B, influenza, and COVID. Finally, I would like to say a few words about treatment withdrawal. You can consider treatment withdrawal if you've been in biochemical remission for at least two years. The first reports of treatment withdrawal were in the setting of performing a liver biopsy, ensuring that you had restoration of liver tissue to normal. And if that was the case, the risk of relapse was as low as 28%. Recently, the need for liver biopsy in adults has been challenged with at least one report stating that treatment withdrawal just based on biochemical remission was enough, and that liver biopsy may not be necessary. In pediatrics, it's strongly recommended that you continue to perform liver biopsies and confirm that the liver histology has a lack of inflammation prior to treatment withdrawal. The largest study that's been reported on this is out of Utah, where they had 35 children in biochemical remission for at least two years. They all underwent a liver biopsy, and 16 of those 35 children, or 46%, were in histologic remission. And of those 16, 14, or 88%, had successful sustained remission after treatment withdrawal. On the right is showing you the overall probability of treatment withdrawal success and sustained remission of approximately 40% for the entire cohort. So, the key takeaways from this talk are that the presentation subtype of autoimmune hepatitis will dictate the treatment plan. The majority of cases of autoimmune hepatitis will respond to standard therapy, steroids for induction and azathioprine for maintenance of remission. You should have a high index of suspicion for concurrent extrahepatic autoimmune diseases with celiac disease and thyroid being the most common. And treatment withdrawal can be attempted in children after you've got proven histologic remission in order to maximize the success of that treatment withdrawal. So, future directions should include research on the underlying immune mechanisms involved in both drug and vaccine-induced autoimmune hepatitis-like injury. Secondly, there is a strong need for prognostic biomarkers that would predict treatment failure, relapse, or progression to cirrhosis. And we also need therapeutic biomarkers that would predict who is in biochemical remission, who is in histologic remission, and if we find an accurate therapeutic biomarker, we may not need a liver biopsy one day. And also, therapeutic biomarkers could be useful as endpoints for novel therapies. Thank you. Hello, everybody. I'd like to thank the organizers for inviting me to talk about challenging cases of autoimmune hepatitis. My name is Saeed Mohammed. I'm the medical director of hepatology and liver transplantation at Northwestern University, and here are my disclosures. For our objectives, I want to talk about autoimmune hepatitis. It's a rare cause of chronic liver disease in children. The majority of children do respond to the standard of care. Some need to be transplanted, especially those who present in acute liver failure. Out of all children, 5 to 10 percent can be very difficult to treat. The options that I'm presenting today are based on the published guidelines and my own experience. So, my topic is difficult to treat. Now, that can mean several things. One thing it can mean is insufficient response. This can be defined as failure to normalize biochemistry in six months. This is a definition from the European Research Network. There's no definitions from the ASLD guidelines. I don't have a cutoff for time, either. It depends how severe the initial presentation was. If the child was listed for a liver transplant and their enzymes are now less than twice the upper limit of normal, I may accept that as a good response. The second thing we got to look at is tolerance. So, this is a little bit more difficult. It occurs in 3 to 5 percent of patients, and it means that the patient is responding to therapy, but the side effects of the medication, usually GI upset or abdominal pain, are so severe that they cause the patient to stop taking the medication. So, this is also a problem. So, this is where the art of medicine really comes into play. When you have a patient like this who doesn't seem to be responding, you have to ask yourself several questions. One of them is to check the dosing. Is the dosing correct for the patient? Dosing, especially in liquids, you may have different concentrations for the prednisone and for imuran. You got to ask carefully about adherence to medications, especially in young adolescents and teenagers. They may stop taking their medications. Do you have the right disease? Now, could this be another disease, such as Wilson's disease, and this may necessitate another biopsy to make sure you are treating the correct illness. Does the patient have any other diseases? So, autoimmune hepatitis being an autoimmune disease, you can have things like celiac disease, you can have rheumatologic diseases, NAFLD, which is the most common liver disease in children now, can also be present and make things difficult. And this is a table from the ASLD guidelines showing definitions of incomplete response and for treatment intolerance. So, they have very good definitions. They just don't have a specified time period that it's supposed to occur. So, I want to look at these guidelines just very briefly. I think they were, it's a fantastic effort and it's one of the most comprehensive guidelines for any disease that I have seen. So, for second-line treatment, they talk about mycophenolate being the best studied and it can also be used as a first-line therapy. They talk about calcineurin inhibitors, both tiprolimus and cyclosporine. Calcineurin inhibitors and mycophenolate are equally efficacious, but mycophenolate has fewer side effects. They also talk about, briefly, sirolimus, infliximab, rituximab, and methotrexate. There was a meta-analysis of second-line therapies done in pediatric patients by Kamath et al. It was published in 2017. You can see that they looked at about 2,000 studies and by the time they sifted through all of them, there were only 15 that were left that could be included in the quantitative synthesis of the meta-analysis. So, there isn't a lot of data on second-line therapies in pediatrics. This shows their findings. You can see at the top, mycophenolate, some patients did respond. The confidence intervals are very wide. The data for tiprolimus is, some responded very well and some didn't respond at all. And again, confidence intervals are very wide. The studies on cyclosporine seemed a little bit more promising, but you can also see that the confidence intervals are just very wide as well. A similar thing was done in adult, in the adult literature. They identified over 1,500 studies and by the time they looked through them, there was 15 studies that they could use for their meta-analysis. In here, you can see on the top left, they looked at tiprolimus and prednisone, then cyclosporine and prednisone, budesonide, and mycophenolate and prednisone. The mycophenolate data is the most numerous, but the best response seemed to be with the tiprolimus and prednisone followed by cyclosporine and prednisone. So, there was a reduction in the tiprolimus groups of 94% had a reduction of their immunotransferases and then 90% in the cyclosporine and prednisone group. I did want to talk a little bit about sirolimus as a rescue therapy. In our center, we've used it quite a bit and we published a series of four patients that were tried on sirolimus. Two out of four improved, one out of four was non-adherent, and one out of four failed. Sirolimus was a nice choice because most of us were familiar about it from the transplant literature. It can expand Tregs, which may help with autoimmune hepatitis because defective Tregs play a role in the disease. It decreases T-cell activation and similar to calcineurin inhibitors, it decreases B-cell stimulatory cytokine mRNA in T-cells leading to less immunoglobulin. So, all these patients had a reduction in their prednisone dosage. You can see from this table that they started off with doses between 0.3 and 0.8 milligrams per kilo per day and ended on more like 0.1 and 0.2. So, I'll now go into some of the case presentations. When treating some of these patients, I feel like you're playing a game of rock, paper, scissors. I think most people know how this goes, but sometimes the rock wins, sometimes scissors wins, sometimes paper wins. It just depends on what you're dealing with and basically who your opponent is. So, my first case was an 18-year-old female who was referred to us after she'd been steroid-dependent for about three years. She was already in a flare when she came to our hospital. We started her on high-dose prednisone and azathioprine. Her numbers did improve with time. But then she had another flare, so mycophenolate was added. The flares continued. Azathioprine was stopped at that time, and tacrolimus was added. She was on the tacrolimus for two or three months, but she had continuous flares. And so, finally, we did add serolimus to the mix and stopped the tacrolimus. She had a liver biopsy at that time that showed grade 4 inflammation and stage 3 fibrosis. So, after seeing that, despite her being on tacrolimus and serolimus and not seeing any improvement, we decided to treat her with rituximab. And this whole process took us about eight or nine months. So, as you can see here from the diagram, her ALT and AST were really quite high in the 1,500 or 1,000 range initially. The little black arrows are where we started dosing her with rituximab. In the first year, she received four doses, and subsequently, she's getting it every six to eight months. And her labs at the most recent check were normal for the first time in about five years. The next case I have was a five-year-old who had morphia, vitiligo, and who was jaundiced. She had an LKM titer of 1 to 10,000. She was started on therapy with azathioprine and prednisone, and she went into remission for about two years. And then, on the advice of her dermatologist, she was switched from azathioprine to mycophenolate because that better treats for morphia. She was stable on this for maybe about two years and then started having flares. So, she was treated with high-dose prednisone, which did not help. She was on it for about six weeks. She had a biopsy, and she had grade two inflammation after six weeks of high-dose prednisone. And at that point, we discussed with the family and with some other colleagues what to treat her with next. And because she had so many other autoimmune diseases, we decided on a course of rituximab for this patient as well. So, you can see here that prior to her first rituximab dose at day zero, she had two significant elevations in her LFTs. And since then, over the last five to six years, she has not had any more flares. Her numbers did not completely go to normal, shown on this graph, but they are near normal. They're less than twice the upper limit. They are near normal. They're less than twice the upper limit. This graph just shows changes in the CD19 count with rituximab over time. So, initially, after the first dose of rituximab, the counts dropped dramatically. And since then, they have remained low. But they do go up almost before each infusion. Her ALT count, though, has stayed normal. The last case I want to talk about is an eight-year-old who came in with jaundice and fatigue. She was diagnosed with autoimmune hepatitis type 2 with stage 3 fibrosis. She responded well to azathioprine and prednisone, but had frequent relapses during the next six years. So, here you can see she was diagnosed in December of 2012, responded very well, had a flare, and we did a biopsy at that time, treated her with steroids or increased her steroids. Over the next five or six years, she had, you can count them, four, three or four more flares. Each of these was really associated with some non-adherence. So, we did a biopsy each time. I did talk to the family and stressed the importance of adherence. This child did have a lot of side effects from the prednisone. I was unable to completely stop. Then, finally, in 2019, she had a flare and her mother thought she wasn't taking her medications. She was biopsied in early 2020, and we had started her on allopurinol at that time because her imirin level was low. That continued for two or three months. Then, the imirin level became too high, so we stopped allopurinol. We tried to decrease her steroids. Her numbers almost normalized, and then her numbers went up again. She was changed from imirin to CELCEPT. However, she couldn't tolerate the CELCEPT, and she was started after that on tiprolimus. For two or three months, her levels varied from anywhere from three to 20. It didn't seem like we could get a stable level with tiprolimus. We took a step back and reassessed whether or not anything else was present. She had an MRCP that was normal. Then, after six months of having really no improvement, we did another biopsy. Then, she was started on rituximab and IVIG. Since starting on the rituximab and IVIG, her numbers have normalized, and she's now doing well. Here's a treatment algorithm proposed by Dr. Bergani for treating patients with autoimmune hepatitis. First, you'll treat with prednisone and azathioprine. If there's no response or frequent relapses, you can switch to mycophenolate. Then, you move on to cyclosporine or tacpro. If there's still no response, then consider rituximab. This is generally what we try to do, but we can't always follow this algorithm with all children. So in conclusion, I want to say that difficult cases are a minority. Most patients do respond very well. You should initially try treating patients with mycophenolate. It's very important to consider medication side effects and concurrent diseases when choosing second and third line therapies. And if you're unsure what to do, please seek the help of your colleagues. I think many people would be very happy to help in these difficult patients. And thank you so much. Hi, everyone. These are my disclosures. Our goals are to review recent literature relevant to pediatric PSC and to incorporate upcoming guidelines and consensus statements into our practice. The key resources we'll use are the upcoming 2021 practice guidelines on PSC that are currently under final review and technically subject to change from the ASLD committee, and also the International PSC Study Group consensus statements on PSC in gastroenterology this year. So we're going to talk through some practice points relevant to diagnosis, large duct, small duct, and especially autoimmune hepatitis overlap, treatment regarding erso, vanco, and azathioprine and steroids, and some cancer screening recommendations. So in regards to diagnosis, the overall algorithm isn't changed much. If you suspect PSC, of course, exclude secondary causes. In kids, this is especially drug-induced liver injury, which is often overlooked, MDR-3 deficiency, and congenital immunodeficiencies. An MRCP is the key imaging test. If there are the typical strictures present, yes, you have PSC. If the findings are equivocal, new in this guidance is to repeat the MRCP in one to two years rather than going to other studies. And if it's entirely normal, consider a liver biopsy to look for small duct PSC. Notably not part of the diagnostic algorithm is ultrasound at all. This is frequently used as a screening test before MRCP, but it lacks the sensitivity to pick up PSC and shouldn't be used and isn't a part of testing for PSC. Similarly, ERCP in the past may have been used for these equivocal or normal cases to distend the ducts and look for strictures, but ERCP should not be used as part of the diagnosis at all. Small duct PSC from the consensus definitions, a little bit more scrutiny here. The big changes are requiring a recently normal MRCP to call it small duct PSC. We'll talk about its progression and how it becomes large duct PSC next. But with those and with excluding other causes, you need either or a classic biopsy, which is basically onion skinning fibrosis or the more common situation, a compatible or nonspecific biopsy down on the right, which is totally nonspecific features of chronic biliary disease, bowel duct loss, duct reaction, chronic cholestasis, none of which are specific for PSC. And so since that's somewhat vague in a lot of cases, the new requirement is that these patients also have inflammatory bowel disease. If they don't, it's just more nebulous and it's not clear exactly what's going on. So a little more scrutiny there. Newer studies on small duct PSC show that over a decade or so, a third to over half of adults end up progressing to large duct PSC and a quarter of children over seven years we showed. So it seems that most or many cases of small duct PSC are simply early large duct PSC that hasn't gotten that far along yet. And so another guidance statement is small duct PSC should be monitored for the development of large duct disease, perhaps every two or three years or with clinical changes. In terms of autoimmune hepatitis, new changes liver biopsy for some and not for all. The 2010 guidance was to biopsy all children to look for autoimmune hepatitis. And in 2020, 2021, liver biopsy should not be performed in patients with typical cholangiographic findings, but should be performed in patients in which there's a concern for overlap. And notably, a concern is not being a child anymore. So where does this take us? It is somewhat vague intentionally. So we just don't have good data to press into who should be biopsied or not. I'll go over that in a second. But there's no clinical or biochemical criteria for that concern. Autoantibody titers and IgG are very difficult to interpret. The consensus definitions separate document suggests a minimum criteria of an ALT of five times the upper limit of normal, which would be 125 to 250, depending on age of patient and center. So I like around 200 as a minimum. There's after we biopsy no criteria for what defines overlap, which makes this difficult. And this is a key research need in pediatrics, which will be directly addressed in the children PSC study. So I'm looking forward to that. So just in terms of practice with an individual patient, a 16-year-old with large duct PSC and modestly elevated labs and a nonspecific ANA, this patient doesn't specifically need to be biopsied. A couple of abstracts here that you can check out as posters. Number 1986 from our group by JP Stevens and Nitika Gupta at Emory looked at data from the PSC consortium, 400 liver biopsies in children with PSC. And having the pathologist stratify is it's not autoimmune hepatitis. Maybe there's some features or yes, it's it's classic overlap autoimmune hepatitis. We found that up on the top there total IgG did not correlate at all with that. So whether the pathologist called it that or not, same total IgG and similarly down below ALT was the same across groups. So and this was true for a bunch of other predictors on the biopsy itself, only fibrosis stage, not features of autoimmunity were predictive of prognosis. And in regards to treatment, immunosuppression didn't alter outcomes in overlap. And this is more consortium data, this time taking 184 patients with overlap, who got either azathioprine and steroids or observation, and matching them on baseline lab and biopsy features of autoimmunity. So similar IgG, similar degree of interface hepatitis, for instance. And we found no difference in outcomes at all. So on the top there ALT over the first year, you can see that on all both groups went down to a normal level over the first year. And then event-free survival on the bottom, you can see that both groups showed that 20% or so dip over the first year that we expect when patients who are cirrhotic with either of these diagnoses show up and we can't fix them. And then after that, identical progression over time. So who, when and why do we biopsy for suspected overlap? Elevated labs don't seem to predict a pathologist diagnosis of autoimmune hepatitis. Features of overlap on biopsy have no prognostic value. Fibrosis stage is prognostic, but we can estimate that in other ways. In fact, this is another guidance statement that we should use liver stiffness and not liver biopsy for fibrosis staging. And even when we label a patient overlap and treat, we don't seem to change the outcome. So this is hypothesis generating data. This is the biggest area that distinguishes children from adults seemingly. And so these are very important things to look into in the coming years. Important possibilities are that overlap may be grossly overdiagnosed in children. Formal diagnostic criteria are definitely needed. Interface activity in plasma cells may be a normal part of PSC. Azathioprine and steroids may be the wrong drugs for overlap. And this is a key research need. So on to treatment. In short summary, erso is a maybe, vanco is a no. This comes from our work, in large part, on these two drugs compared to observation. Notably, I think all six speakers here are on this paper, which I thought was interesting. So 264 matched children with PSC and intention to treat analysis with one year of therapy or observation. And we found no difference in biochemistry, histopathology, or clinical outcomes with either of these therapies. But more important than showing that vanco and erso didn't seem to work was that the perceived response rate, the reduction in GGT as a primary outcome, was really a function of how sick they were at baseline, how severe their disease was, what stage they were in. And as estimated by the scope index of low, medium, or high, to start, you can see that their perceived response was higher in low-risk patients and went down precipitously with more risk. And this even stratified out over the individual numbers, 0 to 11 in the scope index. And with a scope of zero, which is actually many children that we see, they had almost 100% biochemical normalization rate, regardless of therapy. So if a patient is really well at baseline, they're probably going to do well and show a GGT reduction. This is a schema of progression of PSC, a sort of cartoon version, the way that I think about it in an early stage with relapsing and remitting inflammation can see the GGT going up and down. A next stage with progressive continuous inflammation that may not occur for decades after diagnosis, at least years, and then an end stage of fibrosis and obstructed bile flow. And we know that symptoms don't happen until late. They're an end stage complication in most patients. And so we can't really be guided by those. Fibrosis tracks with symptoms and progresses rapidly much later and doesn't really progress much in the early stages. And the challenge is that up to 90% of kids that we take care of are in sort of this window in the box here where inflammatory markers will go up and down. Symptoms don't really happen and don't progress and fibrosis is minimal and also doesn't progress over the time that we take care of them. And most of these kids are very low risk scope. What does that mean? So by individual scope index outcome points, you can see their probability of events and at five years in these low risk patients, they have a negligible, almost non-existent progression to complications. So not a lot is going to happen in these patients. And so as this comes out to treatment, so or so if you really want to try, but consider a waiting period that the key point in this guidance is persistently elevated GGT. Because of this normalization, you really want to know if you're treating a spontaneous normalization with ongoing therapy or if it's actually the effect of your drug. And you can continue if the GGT went down substantially or their symptoms went away if they had any. Vanco, keep it within clinical trials. Antibiotics are for specific indications and not for treatment of the disease. So acute ascending cholangitis, procedural prophylaxis in clinical trials. So can or should you wait to treat? Another guidance statement is to use risk stratification tools. And for kids, that's basically the scope to stage patients and help make treatment decisions. And this would be how that lines up. So can or should you wait for ERSO, Vanco or azathioprine and steroids? There's a negligible risk of progression regardless if you have a low risk scope patient. So you can really step back from treatment. Does my patient really need this? Should I use an off-label or unproven therapy in this patient? Maybe not. There's unlikely that anything is going to happen to them. This is also true of small duck PSC. The largest data shows no clear benefit. Spontaneous normalization is in 50% of children overall and 60 to 100% of children with low risk scope index. So what are we following? We don't really even know. How do you know they're getting better? Very few children have symptoms. Families won't let go of meds once they're started, which complicates care and clinical trials, which is what we really need to do. So wait six months after diagnosis and see what biochemistry does spontaneously before you treat, I think is a good overall guideline. And lastly, cancer screening. So new in this guidance, cholangiocarcinoma screening specifically is not recommended before age 18 or in small duck PSC. It just occurs too rarely. But colorectal dysplasia is a yes, this is new for PSC and IBD patients annual or biannual screening starting at 15. So it's going to be a lot more scopes. And this is based on our data showing that the rate of cancer dysplasia in these patients is 280 per 100,000 patient years and compared to below rates in the general public. So this is 300 times their age matched peers, and younger people on a rate that's similar to senior citizens over 60 to 65 years old. So more scrutiny on these cancer cases. So surveillance colonoscopy starting at age 15 and repeated at one to two year intervals for colonic dysplasia. And if you find biopsy proven invisible, low grade colonic dysplasia, chromoendoscopy is required to follow those patients further. I think this is a key area to potentially engage adult trained GI providers, similar to how ERCP is done at most of our centers. A lot of us just don't have a ton of experience finding these small flat lesions or really even doing screening colonoscopies for cancer. And this is the bread and butter of adult GI. So if you've never heard of chromoendoscopy or don't know what biopsy proven invisible low grade colonic dysplasia is, you should probably be sending the patient to someone else to do these screens. Just a consideration. And lastly, one thing that we can do and advise our patients separate from the guidance. This is a big study in gut, recently showing that sugar sweetened beverages are an increased risk factor for early onset colorectal cancer. So this is reason number 1000 to avoid these. Each additional serving per day between age 13 and 18 imparts a one third higher risk of early onset colorectal cancer. So advise your patients to stop. Thanks so much. That's all I have. I hope this was helpful. Good afternoon, everybody. My name is Amanda Ricciuto. I'm a pediatric gastroenterologist at SickKids in Toronto. And today I'll be talking about modeling prognosis in pediatric PSC. I have no disclosures. So we'll get started. Before talking about prognosis, I think it's important to understand a little bit about the natural history of pediatric PSC. And our best resource to do so to date is the Pediatric PSC Consortium, a huge international registry. In this cohort, liver transplant presurvival was 88% at five years, and 70% at 10 years, event presurvival was 70% of five years. And by 10 years after diagnosis, almost half of children had had experienced an adverse event. Importantly, portal hypertensive complications such as ascites or esophageal varices and biliary complications such as strict urinating intervention were important herald events. The median time delivered transplant after these complications was 2.8 and 3.5 years respectively. In fact, almost three quarters of patients with one of these complications subsequently needed to be transplanted. So although not conventional prognostic markers, they are important herald events. In the same cohort, small duct PSC and concomitant IBD were both independently associated with a decreased risk of progressing to adverse outcomes. PSC AIH overlap, on the other hand, had similar outcomes to classic PSC. If we look to the adult literature, the International PSC Study Group has published a large retrospective study including 7,000 adult PSC patients. They too found that small duct PSC was associated with a lower rate of progressing to liver transplant or death compared to classic PSC and also compared to classic PSC and there was no difference in risk between patients with AIH overlap or classic PSC. In terms of IBD, Crohn's disease was protected. However, ulcerative colitis or IBD-U was not. Looking at malignancy in the pediatric PSC consortium, eight patients developed cholangiocarcinoma. They were all older teenagers and interestingly, all of them had classic PSC. None had PSC AIH overlap. There was also a slight male predominance compared to the overall cohort. Again, looking to the adult literature, we find that there too, classic PSC was associated with an increased risk of hepatobiliary malignancy compared to AIH overlap. In this adult cohort, small duct PSC was also protective against cancer risk. In terms of IBD, having either ulcerative colitis or IBD-U was associated with an increased risk of cancer compared to Crohn's disease or no IBD, although that wasn't seen in the pediatric cohort. But like in the pediatric cohort, males also had an increased risk of malignancy. The question of whether PSC with AIH overlap is associated with different outcomes than PSC is a bit of a hot topic these days. This is a summary of some of the studies, all small in size, that have commented on outcomes in PSC AIH overlap. I won't go through them all individually, but some of the take-home messages are that in PSC with AIH overlap, biochemical and inflammatory histologic improvement are common and occur at rates similar to AIH. However, looking at the outcome of liver transplant resurvival, that tends to be worse in PSC AIH overlap compared to AIH alone, and similar to in PSC cohorts. This leads to the notion then that biochemical and histologic inflammatory improvement in PSC AIH overlap likely do not have the same prognostic significance as they do in isolated autoimmune hepatitis. Moving on to serum markers and liver biopsy. Looking at GGT at time of PSC diagnosis in the pediatric PSC consortium, patients with lower GGT and the lowest tertile had a slower rate of progression to adverse events. This was in contrast to alkaline phosphatase, which was not able to discriminate risk groups, which fits with the idea that alkaline phosphatase is not as useful in pediatrics as it is in adults because alkaline phosphatase fluctuates with age and bone growth. GGT is also a useful marker at one year, specifically normalization to levels less than 50 at one year is associated with a much slower rate of progression to adverse outcomes. Moreover, patients who have the largest decreases in GGT, specifically at least 75%, do best. Interestingly, GGT normalization is associated with better outcomes, regardless of whether or not patients are treated with erso deoxycholic acid. This suggests that those patients who normalize their biochemistry on erso are those who were not destined to experience adverse outcomes anyways. Serum fibrosis tests are employed less commonly because they're not always routinely available, but these are some of the more commonly talked about ones. Enhanced liver fibrosis, or ELF, which includes hyaluronic acid, P3NP, and TMP1, FIP4, APRI, which is the AST to platelet ratio, and serum MMP7 or metalloprotein E7. All of these, in adults, plus or minus pediatrics, have been associated with clinical outcomes in PSC and have been correlated with markers of fibrosis such as elastography or biopsy findings. The APRI score, in particular, has been studied in the Pediatric PSC Consortium, and it was found that patients in the highest APRI tertile experienced a more rapid progression to adverse outcomes. Serum MMP7 distinguishes PSC from autoimmune hepatitis in a pediatric cohort, and it also correlates with findings on liver biopsy and on MRCP. In terms of liver biopsy, there are three scoring systems that have been studied in PSC, the Nakanuma, Eschach, and Ludwig scores. All have been found to be independent predictors of liver transplant and liver-related events. Of course, the issue with liver biopsy is sampling variability given the patchy nature of PSC. Ultimately, liver biopsy is not recommended for routine PSC staging outside of the context of clinical trials. Moving on to elastography, starting with transient elastography. This is a bit of an older study from about seven years ago, but it was a very nice study in that it was prospective and all patients underwent liver biopsy. In the left figure, you can see the very nice stepwise increase in liver stiffness measurements as you move up in fibrosis stage on liver biopsy. Specific cutoffs for F3 and F4 fibrosis stages were 9.6 and 14.4 respectively, both of which performed very well with areas under the curve greater than 0.9. In a separate study, the optimal transient elastography cutoff to predict survival without liver transplant or cirrhotic decompensation was 10.5 kilopascals. MR elastography can also be used to estimate liver fibrosis. It's been studied in a couple of adult cohorts. In two studies, the optimal cut point for F4 fibrosis was found to range between four and five, and cut points for low, medium, and high-risk groups for hepatic decompensation were found to be less than 4.5, 4.5-6, and greater than six. Changes in elastography tend to occur very slowly in the early stages of PSC, but then exponentially faster as the disease progresses to more advanced fibrosis and cirrhosis. In a large retrospective adult PSC study, the mean change in liver stiffness per year was found to be just 0.05 kilopascals per year. The cut point that best predicted hepatic decompensation in terms of change per year was a change of 0.34 kilopascals. Moving on to imaging. The ANALI score is an MRI score that was devised a few years ago. There are two versions, one without gadolidium, which includes dilatation of intrahepatic ducts, dysmorphia, and portal hypertension, and another version with gadolidium, which includes dysmorphia and parenchymal enhancement heterogeneity. The cutoffs to separate high and low risk are respectively two and one. In the original derivation cohort, the scores were assessed for their ability to predict radiologic worsening on repeat MRCP, and they performed quite well with areas under the curve of about 0.8. They were subsequently studied in external cohorts for their ability to predict survival without liver transplant or cirrhotic decompensation, and particularly, the version without gadolidium performed well with a C-statistic of 0.89. In a separate study, the combination of the ANALI score without gadolidium and liver stiffness by FibroScan was studied. The two metrics are found to be correlated, but to be complementary, and together, they nicely defined low, medium, and high-risk groups, as shown in the figure to the right, which illustrates the queer separation and risk based on the combined tools. Looking at cholangiography, back in 2002, a group of investigators devised a classification system for ERCP, which involves staging the degree of abnormalities of the intrahepatic ducts and extrahepatic ducts and then combining the two to generate a sum score. This is then incorporated with patient age to produce a final score. This ERCP score was subsequently validated in an external adult PSC cohort in Oslo and you can see in this figure, the very nice separation of patients in terms of their risk of progressing to adverse outcomes. At our pediatric center, we took a similar classification system, but applied it to MRCP. We similarly summed the intrahepatic and extrahepatic scores, looking at the worst region in both cases. We found that this also performed well to separate patients according to their risk of progressing to PSC complications or liver transplant. As shown in the table below, this MRCP score outperformed all of the other metrics that we examined, including biochemistry, as well as liver fibrosis stage on biopsy and Mayo risk score. Finally, a quick word about clinical risk scores. A number of adult risk scores have been developed over the years, including the Mayo model, more recently the revised version, the Amsterdam-Oxford model, the Boberg model. Then more recently, the UK PSC risk scores, of which there's a short-term and long-term version, and also the PRESTO score, which was developed using machine learning. Most of these tools include items like bilirubin, albumin. Several of them include platelets or liver enzymes, and some of them include age. It's important to take note of the timeframe over which these models are intended to estimate risk, because some are meant to be used just for the short-term estimation of risk, like the Mayo model, and others are intended for longer-term estimation, such as the Amsterdam-Oxford model. All of these tools perform reasonably well in the adult cohorts in which they were studied. Here we see the performance of these adult scores. When applied to a pediatric population, specifically the pediatric PSC consortium population. Here we see statistics graphed over time relative to PSC diagnosis. In both figures, the top line corresponds to the modified Mayo score. Of the adult tools, it performs the best in a pediatric population. But you can appreciate that most of them have a diminishing performance over time. In addition, as a general statement, most of these models tend to underestimate survival in children who are classified as high-risk. This might be due to over-weighting of AST because of the higher prevalence of AIH overlap in pediatric populations compared to adults. The imperfect performance of the adult tools led the Pediatric PSC Consortium Group to develop the first pediatric PSC clinical risk score, the SCOPE index. It was developed in a very large international cohort of 1,000 patients against a primary endpoint of liver transplant or death. It was internally validated, it has yet to be externally validated, and it estimates risk at 1, 5, and 10 years. It has good discriminative ability at each of those time points as highlighted by C statistics of around 0.8 or higher. It also performs well when using data at two years from diagnosis instead of at time of diagnosis, and it performs similarly regardless of disease phenotype. The items that it includes are total bilirubin, albumin, platelet count, GGT, and whether or not cholangiography is abnormal. The cutoffs for low, medium, and high-risk are respectively 0-3, 4-5, and 6-11. You can see that patients in the low-risk group have a very low risk of the primary outcome every year, just 1 percent. In the high-risk group, that risk increases to 9 percent per year. When we look at the performance of the scope index compared to the adult models, we see that it outperforms all of the adult risk tools regardless of whether using data from time of diagnosis or at two years. This has led to the recommendation from AASLD that in children the scope index be used as a clinical risk score of choice, except in the setting of end-stage liver disease, in which case the PEL should be used. A few take-home points. Staging PSC is inherently tricky because of the variable rate of progression of disease across patients. Nevertheless, risk stratification and fibrosis staging should be attempted at diagnosis and over follow-up. Liver stiffness measurement is preferred to estimate fibrosis stage over biopsy, which is not routinely recommended for that purpose. Clinical risk scores like the scope index can be used, but one should keep in mind that the estimated probabilities are generalizations and they may not apply perfectly to the patient in front of you. Then finally, a quick plug about the children PSC cohort study, which is in its early phases. This prospective multicenter cohort will allow us to study prognostic markers like the ones I've talked about and others, and we'll add very meaningfully to this field. Thank you very much for your attention. Good morning. I want to thank the organizer for the opportunity to present today, and all of you for taking time out of your busiest case to join us in this discussion about autoimmune disease recurrent impact on patient and allograft outcome. Mercedes Martinez, Epidiatric Transplant Hepatology at New York Presbyterian. I have no disclosures. The objective of this presentation is to review the diagnostic criteria for recurrent autoimmune hepatitis and primary sclerosis, cholangitis, when I refer from now on as RAIH and RPSC. Assess the risk factors for autoimmune recurrence in the allograft, and review the management and impact of RAIH into the liver allograft and the morbidity and mortality due to RPSC. Liver transplantation is the only treatment modality when autoimmune liver diseases progress to an stage liver disease. The need of transplantation have decreased over the years, mainly for autoimmune hepatitis due to availability of effective medical treatment. The long-term outcomes are affected by disease recurrence mainly in RPSC. Recurrent disease may be silent, only distinguishable from allograft rejection, which makes sometimes patients presented when the allograft has been irreversible damage. Then let's focus on recurrent autoimmune hepatitis. The diagnosis criteria is based on a combination of clinical, biochemical, immunological, and histological, similar to the criteria used for the diagnosis of AIH in non-transplant setting. Although some features may be less pronounced or absent because of the concurrent use of immunotherapy. Clinically, AIH was the indication for liver transplant. Biochemical, when gastric function present, usually has hepatocellular injury, predominantly. Patient has hypergamma globulinemia and autoantibodies, like the ones used for the diagnosis of AIH. And a compatible histology that present with interphase hepatitis, portal inflammation of lymphoplasmositic inflammatory infiltrates, and the development of rosettes. Very important, this patient responds to corticosteroid therapy. But it will be very important that we exclude other cause of allograft dysfunction, which in the setting of transplant is acute cellular rejection. When we look at the risk factors reported in the literature for AIH, there are several. Almost everybody agreed that poorly controlled AIH brought to transplant increased the risk of recurrence. And this is presented as high transaminitis and immunoglobulins prior to transplant, or necroinflammatory activity in the X-plan. Also, patients with coexisting autoimmune diseases, or a strong association with HLA-DR3 and DR4, and the rapid winning of a steroid post-transplant. Very many providers have this patient with a slow steroid taper on a steroid for a prolonged period of time. There are other risk factors, which everybody doesn't agree on those, but those are incident increased rates of rejection, the degree of donor-recipient HLA mismatching, and the type of immunosuppression maintenance, whether it's cyclosporine or ticrolimus. Most patients with RAIH responds well to intensified immunosuppressive therapy, whether it's the reintroduction or increased dose of corticosteroid, or adding other immunosuppressive agents like esatoprine, MMF, or rapamune. We almost all agree that if you are increasing or adding new immunosuppressants, you might lower the doses of CNI. The long-term outcome doesn't appear to be impaired in the vast majority of these patients, since less than 5% of patients transplanted for AIH require re-transplant due to disease recurrence. There are several small centers and adult studies reporting on these outcomes, but I think that the best report that I found is this very elegant evaluation of the outcome in children who underwent liver transplantation for autoimmune hepatitis and reported to the SPLIT database. This report was by Martin and collaborators. They report here on 113 patients transplanted for AIH and compared them with a cohort of more than 2,000 patients transplanted for other etiology. And what they demonstrated is that the patient at allograft survival at five years was very similar between these two cohorts. They further evaluate the impact of rejection, and when the patients with autoimmune hepatitis have maybe a mild increase of cellular rejection compared with the other cohort, at five years, this difference was not significant. Let's move to AIH. The diagnostic criteria for PSC, I'm sorry, recurrent PSC, is diagnosed with a GRAS-CD or major criteria. Patients underwent PSC with a primary indication of transplant, and this diagnosis shouldn't be made in the first 90 days. Patients present with allograft dysfunction, usually a cholestatic hepatitis, and one or more of the following criteria. Typical histological finding of fibro-orbital lesion around the bile ducts, or a cholangiography demonstrating multifocal non-anastomotic biliary restriction with bleeding of the bile duct, and this could be an MRCP, an ERCP, or PTC. Very important, we should exclude other causes of GRAS dysfunction, like patient with chronic ductal penis rejection, or patient with significant bile duct damage due to vascular events like hepatic artery thrombosis. The risk factor of PSC are many, this is just a selection of them. Certain HLAs are associated with recipient or donor, could predispose to recurrent male recipient or recipient donor's gender mismatch. Recipient of donor age, whether it's too old or too young, extended donor criteria, allograft, and the presence of acute cellular rejection that is difficult to treat and require the use of depleting therapy. I think that almost everybody agree upon that IBD and the presence of an intact colon and IBD activity are risk factors for recurrent PSC. In this very elegant report of the SPLIT database, Miller and collaborator reported that the outcome of 79 patients with PSC have a similar outcome at five years with 378 patients without PSC. Then this was about a decade ago. More recently, we have reported in the experience of the PSC database, which is the multi-center cohort. At the time of our evaluation, we reported in the transplant activity of 1,300 children. This was recently published in Hepatology. About this more than 1,300 patients, 140 patients of our cohort have undergone liver transplantation. And at the time of the data enter, more than 90 days has passed from the transplant. This represent 592 person years follow-up. 55% were male. Patients in this cohort were half a million age at diagnosis of 12 year, 15.4 years at the time of transplant, and there were less than three years that passed between diagnosis and listing. Then 36 out of these 140 patients have recurrent PSC with 27% having recurrent PSC at five years. As we can demonstrate in this graph, the recurrent PSC increase as a function of the time after transplant. We try to evaluate certain risk factors or potential risk factors for recurrent PSC. When we look at the demographics, we found that the cohort with recurrent PSC was significantly younger at transplantation, and the years from transplant to diagnosis was significantly shorter. When we look at the association with other autoimmune diseases, we found that the cohort with recurrent PSC has a significantly higher percentage of IBD. And also there were an increased number of patient with AIH, though this difference didn't achieve a significant difference. IBD remission or type of IBD didn't achieve a statistical significance. When we look at the biochemical values, a liver transplant, we found that patients with recurrent PSC have a significant value of ALT at time of transplantation, but no difference regarding bilirubin, albumin, INR, GGT, or blood cell counts. We evaluate several peritransplant variables, including type of graft, type of donor, donor age, ischemia time, and type of anastomosis, and we found no significant difference, though the majority of the patient have a Rheumatoid Arthritis on both cohorts. When we evaluate the induction and maintenance medication, we found no significant difference regarding the induction or maintenance immunosuppression, and also no significant difference of the use of also the oxycolic acid post-liver transplant in either cohort. When we evaluate the episode of rejection, we found that patients with recurrent PSC have significant higher number of episode of rejection, significant number of episode of rejection per year, and the percentage of esteroid refractory rejection was also significantly higher in the recurrent PSC cohort when compared with a non-recurrent PSC cohort. The allograft survival was worse in the recurrent PSC cohort. What we see here in this curve represented three deaths in the patients with non-recurrent PSC, one from progressive cholangiocarcinoma, and two from sepsis. In the recurrent PSC cohort, there were six patients released before transplant, four of which received a transplant, one patient died before transplant of end-stage liver disease, one died of sepsis, and two died of peritransplant complications. Three patients underwent successful liver transplant and are alive. Also, we demonstrate here that recurrent PSC increased morbidity. We report here on events related to portal hypertension listed for re-transplant or death for liver disease, and we see that 25% of the patients have achieved some of these endpoints at one year and 43% at two year. One recurrent PSC comes, it's highly morbid for these patients. Also, the allograft survival after liver transplant for recurrent PSC is affected. Only 90% at one year, 81% at two years, and 63% at five years, which is significantly lower than the percentage for patients that are not transplanted due to PSC, 95, 93, and 88% at five years. In conclusion, recurrent autoimmunity increase with time, and the diagnosis is highly variable depending on the centers. Most risk factors described are based on adult single-center small cohorts. Then we as pediatricians and advocates for children should try to study this more in detail in larger multi-center cohorts. Our AIH respond to medical management and those in significant impact patient or allograft survival, but our PSC is associated with substantial morbidity and mortality. I think that I'm going to leave you with these future directions. We really should aim to do prospective research with collection of clinical data and sample collection for immunophenotyping. We should uncover the pathophysiology of PSC before transplant and recurring PSC using newly innovated methods and develop new therapies or immunosuppression strategies to prevent and treat recurrent autoimmunity after liver transplant. I also want to recognize the PSC consortium, mark the note that have done an outstanding job in these conditions, and I will be happy to answer any questions. Hi, everyone. I'm Amber Hildreth from UC San Diego. Just wanted to thank you all for attending our session today. Also wanted to extend a huge thank you to our speakers for providing awesome presentations. If you have any comments or any suggestions for future ASLB or NASPGAN Pediatric Symposium topics, please feel free to reach out to Mercedes Martinez or Saeed Mohammed. Otherwise, I hope you all enjoy the rest of the liver meeting, and of course, hope that we all get to see each other in person next year.

2021 LIBER Meeting

pediatric autoimmune liver diseases

autoimmune hepatitis

primary sclerosing cholangitis

pathophysiology

diagnosis

management strategies

genetic predisposition

environmental triggers

medications

treatment challenges

collaboration