Welcome to the Cholestasis and Autoimmune Liver Diseases Special Interest Group. On behalf of the program chairs and moderators, we would like to welcome you to this special interest group. This year, the focus has been on a disease where knowledge has been lacking and with limited options in treatment. The topics have been chosen to highlight advances and progress made in diagnosis and management. Each presentation will be 15 minutes, and we will be answering any questions that you may have in the chat section. Here you will see a list of the upcoming talks. And with that, we will kick things off with the gut microbiome and fecal microbiota transplantation for the treatment of primary sclerosing cholangitis. My name is Jessica Allegretti, and I am the director of the Fecal Microbiota Transplant Program at the Brigham and Women's Hospital in Boston, Massachusetts. These are my relevant disclosures. It's important to start with some key definitions and concepts. Microbiota is essentially the collection of microbial species that form a microbial community. The microbiome is the collection of genetic material present in the genomes of microorganisms present in that committee. Dysbiosis essentially means a derangement of that microbial community. And commensals are microorganisms that live in intimate contact with another and derive benefit without harming or benefit the other. In health, generally speaking, there are several types of microorganisms present in the human body, bacteria only being one of them. This includes archaea, viruses, fungi, and even some protozoa. Among the bacteria, however, this is really dominated by four major phyla, the firmicutes, bacteroidetes, proteobacteria, and actinobacteria. Our microbiomes generally are finalized, if you will, by age three, and this is often dramatically influenced by how you're delivered, your initial environmental exposures, and your early life diet. The abundant intestinal flora is tolerated through a complex host of immune interactions, and disruption to this balance has been implicated in the pathogenesis of several infectious as well as inflammatory GI diseases. And so it's easy to imagine that many things can go wrong. We know that a dysbiosis disrupts health, and there have been many dysbiosis-related diseases that have been identified, including C. diff infection, antibiotic-associated diarrhea, and inflammatory bowel disease. One problem is that antibiotics continue to occupy an important part of the therapeutic management in most of these diseases, and ongoing loss of bacterial diversity and incomplete microbial recovery really continue to lend itself to the problem. And so, hence, restoration of a healthy intestinal flora through manipulation of the gut microbiome has gained a lot of interest in the treatment of many gastrointestinal diseases. And so many have turned to fecal microbiota transplantation, or FMT, and it has become one of the most commonly utilized ways to restore gut microbial composition as well as functionality. So what is fecal microbiota transplantation? It is the installation of minimally manipulated microbial communities from stool of a healthy donor into a patient's GI tract. And that's really how FMT is distinguished from a defined consortium of microorganisms, highlighting the degree of complexity and functionality of the microbiome. And it is because of this complexity that the FDA considers this therapy to be both a drug as well as a biologic agent. This therapy is currently not FDA approved for any indication, but under a policy called enforcement discretion, clinicians may use this to treat specifically C. diff not responding to standard therapy. Now this policy states that you do not need an investigational new drug license in order to use this therapy specifically for C. diff. However, you do have to state that the therapy is investigational and discuss the real and theoretical risks. For all other indications, including PSC, which we're going to discuss, an IND would be required, and this must be done in the setting of a clinical trial. I do think it's important to also discuss the recent safety concerns that have been brought to light with a few cases that involve FMT. Specifically last summer, there were MDRO transmissions that were reported by the FDA. Specifically two immunocompromised patients developed a systemic infection with an extended spectrum beta-lactamase producing E. coli after FMT, and unfortunately one of these patients did die. What's important to know about these cases is that these were done under an IND approved protocol in clinical trials for indications other than C. diff. One was for GVHD and the other for hepatic encephalopathy, and this was not done using stool banked material but an individual hospital's internal stool bank. And I think scenarios like this really emphasize the importance of meticulous and standardized donor screening for potential pathogens. We have to, of course, mention COVID-19 as well. American Biome, which is the largest stool bank in this country, has already implemented SARS-CoV-2 screening in asymptomatic donors with nasal swabs. Stool testing is also being explored and protocols are currently being implemented. And to date, there have been no reports of transmission of SARS-CoV-2 via FMT. Lastly, there have been reports of EPEC and STEC, two strains of E. coli that have also been recently reported. To note, stool has been screened for EHEC and STEC via enzyme immunoassay traditionally. EPEC had actually not been considered a pathogen, it still isn't really, and in Europe they're actually not screening for this organism in their stool donor protocols. Four cases of STEC and two of EPEC had been reported post-FMT. All of these cases resulted in self-limiting diarrhea. However, because of these reports, STEC and EPEC are now being tested via PCR and this has been implemented across all open biome donors. Irregardless though, FMT does have a very well-established role in the treatment of recurrent C. diff. Here you can see several positive trials where FMT has been administered using various delivery modalities and this is ID and GI guideline approved in both the US and Europe. So now let's turn our attention to primary sclerosing cholangitis, which I don't need to tell this audience, is a progressive chronic cholestatic liver disease that has inflammatory and fibrotic destruction of the bile ducts. It will progress to cirrhosis and often require a liver transplant and most patients have concurrent inflammatory bowel disease and unfortunately there currently are no effective medical therapies for this disease. The pathogenesis of this disease remains largely unknown. There are several hypotheses, however, one that bacterial components may stimulate an aberrant immune response resulting in the perpetuation of the biliary inflammation seen in PSC and it's also thought that bacteria may gain access to the liver and biliary tree through translocation across an abnormal and inflamed intestinal mucosa into the portal venous system. It has been shown that PSC patients have a dysbiotic gut microbiota and as you can see here when you compare patients with PSC to healthy controls, PSC patients certainly have lower biodiversity than healthy controls and specifically when patients with both PSC and IBD were compared to IBD patients without PSC and healthy controls, you can see in this plot that all three of these different patient types have very different gut microbiomes and so patients with PSC are a distinct disease phenotype compared to patients with IBD alone. To begin to explore this, our group had previously assessed the bile of patients with PSC compared to those without PSC and just on culture, we found that the composition of the bile looked quite different whereas in non-PSC patients, you see basically gut contents whereas in PSC patients, you see either no growth at all and really or really a strong signal for strep viridans. Similarly, when we performed 16S sequencing on the bile of these patients, you see overall significantly lower biodiversity in patients with PSC compared to non-PSC patients. This led us to hypothesize that fecal microbiota transplantation will correct a dysbiosis that has led to hepatobiliary inflammation seen in PSC and will therefore improve LFTs and slow progression to cirrhosis. We performed an open-label pilot study of 10 patients with PSC as well as co-occurring inflammatory bowel disease. All patients received a single FMT via colonoscopy from the same unique stool donor. Blood and stool was collected prior to FMT and through week 24. Blood was tested for liver function tests at each time point as well as inflammatory markers. Fecal microbiome analyses were performed using 16S rRNA sequencing at each time point as well as bile acid metabolomic profiling. We included adult patients with confirmed PSC again as well as concurrent inflammatory bowel disease. Patients had to have elevated LFTs at least 1.5 times the upper limit of normal for entry. We excluded patients with decompensated liver disease, any immunocompromised state, antibiotic use within two months of the start date, and any patients on ursodeoxycholic acid needed to be washed out for at least four weeks prior to their baseline visit. Our primary endpoint was mean change in serum liver biochemistries after six months compared to baseline. And treatment success was defined as an improvement in ALKFOS, ALT, or AST by 50% or greater at any time point. Secondary endpoints included microbial diversity at 12 weeks after FMT compared to baseline. Here's our patient population. Again, we enrolled 10 patients. The vast majority were male, average age around 41. All of the patients had UC except for one patient that had Crohn's colitis. And you can see the baseline ALKFOS was almost 500, mean baseline ALKFOS, I should say. With regards to safety, there were no related adverse events noted. One serious adverse event was documented that was found to be unrelated to the FMT. With regards to the clinical results, three of the 10 patients achieved the primary endpoint of success, which was a 50% decrease in their ALKFOS by 24 weeks. Notably, seven of the 10 patients experienced a 30% decrease in at least one of their biomarkers post-FMT. We did note shifts in their taxa composition, much more similar to the donor, at least one week post-FMT as compared to their baseline assessment. Again, here we see essentially global shifts in the microbiomes of these PSC patients as early as week one. And when this was assessed through week 24, these changes were noted to be sustained even after only a single FMT. We identified 381 engraftor strains. Engraftor strains are defined as strains that are present in the donor, missing in the patients pre-FMT, and found to be present in the patients at least one week post-FMT. We did find that there was a correlation between a decrease in ALKFOS level as well as an abundance in these engraftor strains, meaning that the more engraftor strains present in the patient post-FMT, the more likely they were to have lower ALKFOS levels post-FMT on assessment. In summary, this is the first human trial of FMT for the treatment of PSC. We did find that FMT was very safe in this patient population. We found that there was improvement in overall microbial diversity that did persist through week 24. The abundance of engraftor OTUs present in patients post-FMT did correlate with a decrease in ALKFOS levels. However, of course, we recognize that further studies are needed to determine the utility of isolating those engraftor OTUs as well as to understand the role of the microbiome in the pathophysiology of PSC. Thank you so much for your time and attention. I would like to thank all of my collaborators as well as the patients who participated in this study. Have a great rest of your meeting. Hello. My name is Samara Rizvi. I'm a transplant hepatologist at the Mayo Clinic in Rochester, Minnesota with an interest in hepatobiliary neoplasia and cholestatic liver diseases. And my research efforts are centered on the immunobiology of cholangiocarcinoma. And today, I'll be providing an update on cholangiocarcinoma. I have no disclosures. An overview for my talk today, I'll begin with a background on cholangiocarcinoma or CCA, move on to diagnostic modalities, liver transplantation for PSC-associated CCA, and wrap up with emerging systemic therapies for CCA. CCAs are classified into three subtypes based on their anatomic location within the biliary tree. Intrapathic CCAs or ICCAs are located within the hepatic parenchyma, and they arise above the second-order bile ducts. Perihilar CCAs or PCCAs arise between the second-order bile ducts and the cystic duct insertion. And distal CCAs or DCCAs arise between the cystic duct insertion and the ampulla. Perihilar CCA is a CCA subtype that typically arises in the setting of PSC. Perihilar tumors have a tropism or affinity for bile and initially tend to grow longitudinally along the bile ducts. And this makes the diagnosis of early perihilar tumors quite challenging, as imaging studies such as MRI or CT do not show a mass lesion at this early stage. As a tumor advances, radial growth away from the bile duct occurs, and this results in formation of a mass lesion, which may be detected on cross-sectional imaging. As you're all well aware, PSC is an important risk factor for CCA. Approximately a quarter of PSC patients are diagnosed with CCA within the first year of PSC diagnosis, 37% in the first 10 years, and another 37% are diagnosed with CCA 10 or more years after PSC diagnosis. And the lifetime risk of CCA in PSC is approximately 10% to 15%, with an annual incidence of 1% to 2%. And CCA accounts for one-third of all PSC-related deaths. MRI-MRCP has superior sensitivity and specificity for CCA detection compared to other modalities such as ultrasound or PET scan, hence the surveillance strategy of choice in PSC is MRI-MRCP. Moreover, Dr. Eaton and colleagues have recently published their work comparing the diagnostic performance of ultrasound versus MRI-based detection of CCA in PSC patients. As you can see in the top figure, patients with asymptomatic CCA detected only on MRI had an improved five-year survival and progression-free survival compared to those with asymptomatic CCA detected on ultrasound. Moreover, patients who had annual imaging surveillance prior to their diagnosis of CCA had a reduction in mortality as well as CCA progression compared to those who did not have annual imaging surveillance. These results suggest that MRI is superior to ultrasound for early-stage CCA detection in PSC patients and have important implications for our practice. ERC plays an important role in the diagnosis of CCA as it allows for the detection of malignant strictures as well as acquisition of biliary brushings for cytologic diagnosis. Atypical cytology is frequently encountered in PSC and by itself should not raise concern. This is primarily due to the presence of biliary inflammation. Suspicious cytology does represent a concerning finding. Approximately 30% to 40% of PSC patients without a mass lesion and suspicious cytology may ultimately be diagnosed with CCA. Moreover, suspicious cytology is an independent predictor for the development of biliary cancer. And positive cytology is diagnostic of CCA. The other cytologic technique employed in the diagnosis of perihiloclangic carcinoma is fluorescence in situ hybridization or FISH analysis. This method uses fluorescently labeled DNA probes that hybridize to four chromosomal locations. FISH results can be categorized as follows, normal or disomy or polysomy, which is a presence of four or more epithelial cells with two or more signals in two or more of these loci. FISH polysomy indicating duplication of more than one chromosome is a marker for chromosomal instability, which is a hallmark of cancer. And FISH has enhanced sensitivity and similar specificity for CCA diagnosis compared to conventional cytology. Emerging diagnostic modalities for CCA include circulating tumor cells, cell-free DNA, next generation sequencing, amongst others. And we're also at the Mayo Clinic evaluating droplet PCR. And the preliminary data indicate that this may actually be better than FISH. Moving on to the management of clangiocarcinoma. Surgical resection is typically not considered an option for CCA arising in PSC due to the presence of a field defect and underlying parenchymal liver disease. Clangioadjuvant chemoradiation plus liver transplantation is a treatment of choice for the subset of patients with early stage periheliclangiocarcinoma. And the criteria for this protocol include the presence. If there is a mass, then the radial diameter should be less than 3 centimeters. There is no cutoff or longitudinal diameter and absence of metastases in a patient who is otherwise a good candidate for liver transplantation. The treatment protocol includes external beam radiation therapy along with radiation sensitizing chemotherapy with 5-fluorouracil followed by brachytherapy. And then patients are placed on a maintenance capecitabine, which they remain on until closer to the time of transplant, at which point they undergo hand-assisted laparoscopic staging. And if this is negative, then patients go on to liver transplantation. And at the Mayo Clinic, up until 2018, 211 patients with a PSC-related perihelic CCA have undergone neoadjuvant therapy. And of these, 138 have undergone liver transplantation thus far. As you can see here, the outcomes for these patients are excellent with a five-year survival approaching 75% whether patients undergo living donor transplantation or deceased donor transplantation. And again, these outcomes are for patients undergoing liver transplantation following neoadjuvant chemoradiation for PSC-related perihyalclangiocarcinoma. Moving on to emerging systemic therapies for clangiocarcinoma, I'll provide an overview of targeted therapies and then end with immunotherapy. Each anatomic subtype of CCA has a distinct mutational landscape. So genetic alterations that are seen in intraepatic clangiocarcinoma, such as fibroblast growth factor receptor or FGFR gene fusions, are different than the alterations that are seen in perihyalclangiocarcinoma, which include PRKACA or PRKACB gene fusions. A subset of patients with CCA have targetable genetic mutations, such as isocitrate dehydrogenase mutations, and there are IDH inhibitors, as well as FGFR2 genetic aberrations, for which there are FGFR targeting therapies. And in the next two slides, I will review emerging data for IDH and FGFR targeted therapies. And it's important to note that these genetic aberrations tend to primarily occur in patients with intraepatic rather than perihyal or CCA. IvoSedNIP targets mutant IDH1. IvoSedNIP was evaluated in a phase 3 randomized control trial with a crossover design in chemotherapy refractory CCA patients. The primary endpoint was progression-free survival. And as you can see here, PFS was significantly improved with IvoSedNIP compared with the placebo with an advantage of 2.7 months compared to 1.4 months. Panamagatinib is a potent selective inhibitor of FGFR123. And in a multi-center open label phase 2 trial of patients with locally advanced or metastatic clangiocarcinoma with a primary endpoint of patients with objective response, the majority of patients with FGFR fusions or rearrangements did have a measurable response with panamagatinib. And approximately 35% of patients overall with FGFR2 fusions or rearrangements had an objective response in this trial. And the progression-free survival in patients with FGFR2 fusions or rearrangements was 6.9 months compared to 2.1 months for those who had other FGF or FGFR alterations. CCAs are dense desmoplastic tumors with an abundant tumor microenvironment, which also includes a preponderance of immune cells such as CD8 T cells, natural killer cells, or NK cells, as well as immunosuppressive cells, including regulatory T cells or Tregs, and tumor-associated cancer cells. And in this case, the tumor-associated cancer cells or T cells or Tregs, and tumor-associated macrophages or TAMs. Now, immune checkpoint blockade with agents targeting Programmed Death-1 or PD-1 and cytotoxic T lymphocyte-associated protein-4 or CTLA-4, both of which are immune checkpoints. Agents targeting these immune checkpoints have revolutionized cancer treatment across a variety of malignancies. Emerging data in CCA have demonstrated a modest efficacy of immune checkpoint blockade in CCA patients. Nivolumab is a PD-1 targeting therapy, and in a phase II single-arm trial in patients with biliary tract cancers, including five with either perihyal or distal CCA, Nivolumab in the intent-to-treat population had a median overall survival of 14.2 months. Combination NTPD-1 with Nivolumab and CTLA-4 with the ipilimumab is also being evaluated in biliary tract malignancies. Phase II trial data indicated overall survival of 5.7 months and a progression-free survival of 2.9 months with this combination. Now, although there were 13 patients with either perihyal or distal CCA in this study, the responses were exclusively observed in patients who had either intrapatic laryngeal carcinoma or gallbladder cancer. Now, we know that the different subtypes have distinct molecular pathogenesis, and these results indicate that the response to immunotherapy also varies based on the anatomic location of CCA. Now, it's possible that non-responders to immune checkpoint blockade have a non-T cell infiltrated or an immune cold CCA with an abundance of immunosuppressive elements, such as tumor-associated macrophages or myeloid-derived suppressor cells, or MDSCs, and targeting these populations in combination with immune checkpoint blockade may result in improved response. Indeed, in my lab, we have recently shown that PD-L1-positive TAMs increase in CCA, and these PD-L1-positive TAMs promote exhaustion of cytotoxic T lymphocytes and increased tumor progression. However, inhibition of these TAMs promotes a compensatory emergence of granulocytic myeloid-derived suppressor cells, which is an immunosuppressive cell population. And dual inhibition of GMDSCs with TAMs in our murine model of CCA potentiates immune checkpoint blockade with NTPD-1 with consequent reduction in the CCA tumor burden and improvement in murine survival. And thus, these results suggest that immunotherapies targeting elements of both the adaptive and innate immune system may are a promising approach in CCA. Key takeaways. CCA remains a leading source of mortality in PSC patients. The use of MRI to detect CCA in PSC patients is associated with improved outcomes. Neoadjuvant chemoradiation and liver transplantation has excellent outcomes for PSC-related periheliclangiocarcinoma, and immunotherapies targeting both the adaptive and innate immune response in CCA are needed. Thank you very much for your time. Good afternoon. My name is Cara Mack, and I'll be discussing what is new in pediatric PSC. I'd like to thank the ASLD for inviting me to speak. Here is my biographical information and my conflict of interest. Today, I'll provide an overview of the natural progression of pediatric PSC. I'll discuss recent treatment analyses in pediatric PSC and provide an update on discoveries related to prognostic tools for children with PSC. With regard to the natural progression of disease, much of what we have learned in recent years is due to the efforts of the Pediatric PSC Consortium, a worldwide initiative led by Mark Deneau from Utah that encompasses 46 centers from 16 nations and collects retrospective information on over 1,000 children with PSC. Data from the PSC Consortium has revealed that 70% of children will have survival with their native liver 10 years after diagnosis. The average age at the time of liver transplant was 16.5 years. 40% of these children had evidence of portal hypertension, and 25% had a biliary complication, such as biliary stent placement or hospitalization for cholangitis by 10 years after diagnosis. Further analyses based on the frequency of having any event associated with worse outcome, including portal hypertension, biliary stricture requiring intervention, liver transplant, cholangiocarcinoma, or a liver-related death, revealed that only 53% of children were free from a poor outcome event 10 years after diagnosis. Interestingly, there was no difference in the frequency of poor outcome events between those with PSC-only versus PSC autoimmune hepatitis overlap syndrome. Now, an important biomarker that was associated with the frequency of a poor outcome event was GGT. If the GGT at one year after diagnosis was less than 50, then the frequency of event-free survival was significantly higher compared to those children with a GGT of greater than or equal to 50 one year after diagnosis. And the reason as to why some children normalize their GGT in that first year is not well understood. The Toronto Group, led by Amanda Ricciuto, is also looking at outcomes in pediatric PSC, and this study compared IBD patients with or without PSC. Analysis of growth based on height-for-age Z-scores and weight-for-age Z-scores revealed that children with PSC-IBD had significantly lower Z-scores at presentation and at final follow-up compared to IBD-only children. In contrast to the worse growth parameters, the PSC-IBD group had a milder IBD clinical course compared to the IBD-only group, and this was based on higher rates of event-free survival of time to colectomy and or time to biologics, time to the use of non-induction steroids, and the time to IBD-related hospitalizations, and this was over a 12.5-year timeframe. Now, the discrepancy between the poorer growth yet milder IBD clinical activity in the PSC-IBD group is not well understood. Switching now to treatment analyses for pediatric PSC. A multi-centered group recently published on the effects of the withdrawal and reinstitution of ursodeoxycholic acid in pediatric PSC. It included 27 children who had been on UDCA for at least six months at the time of enrollment, and this was phase one or baseline. They were then weaned off of UDCA over an eight-week time period, and the end of this wean was phase three when they're completely off UDCA, and that was followed by reinitiation of UDCA for another eight weeks to complete phase four. The primary outcome was the change in ALT and GGT between phases one and the end of phase three. The results showed that the patients fell equally into one of three response groups. The null group, whose ALT and GGT did not change between phase one and three, the flare group, whose ALT or GGT rose to greater than 100 between phases, and the indeterminate group, where the values were intermediate between the null and the flare. So this graph shows the primary outcomes of the three groups, and you can see that there's a significant increase in GGT overall for all patients, as well as for the flare and the indeterminate groups. Analysis of the time period from phase three, where you're off UDCA, to phase four after eight weeks of being on UDCA, we found that there were significant decreases in the GGT in the flare group. So this suggested that UDCA may be beneficial in a subgroup of patients, specifically the flare group. However, this was a very small study and a much larger placebo-controlled trial would be necessary to determine if there's any true efficacy to UDCA. The Pediatric PSC Consortium also analyzed the response of UDCA, as well as the response to oral vancomycin, and compared these treatments to children who did not receive either of these therapies. This was a retrospective analysis of 860 patients who had complete pharmacy data. A propensity score was generated in order to match the baseline characteristics between groups. This resulted in 88 subjects matched based on their propensity score in each group, oral vancomycin, UDCA, or no therapy. Analysis of GGT levels revealed that the GGT significantly decreased in all groups at six months, both those taking vanco, UDCA, or no treatment. Furthermore, there was no difference between these groups in the level of improvement of GGT. The decrease in GGT at six months was sustained at the 12-month time point, and there was no difference between the groups. There was also no difference between groups in the frequency of being listed for liver transplant over a five-year time span, and the probability of needing a transplant was approximately 15% for all groups. So this retrospective study would suggest that neither UDCA nor vancomycin had a significant impact on outcomes in pediatric PSC. Finally, I'll be discussing two new prognostic tools for pediatric PSC. First is a study from the Toronto Group looking at an MRCP score as a prognostic indicator in pediatric PSC. The aim was to determine the inter-rater reliability and prognostic utility of the modified Majoi ERCP classification that is applied to MRCPs. This modified classification gives scores for abnormalities found in both the intrahepatic ducts and the extrahepatic biliary tree. They found that if the total score of both the intra- and extrahepatic findings was greater than or equal to four, there was a 70% frequency of portal hypertension, 50% for liver transplant, and 60% frequency of PSC-related complications, significantly higher than in the group that had a score of less than four. The cohort was also stratified into three risk groups based on the MRCP score, and they identified that the individual risk group also predicted frequency of being free from a PSC complication. For example, the low-risk group with scores of zero to one were 100% free from a PSC complication compared to the high-risk group where only 40% were survival-free from a PSC complication. The other exciting prognostic tool that's currently in press is the Sclerosing Cholangitis Outcomes in Pediatrics Index, or SCOPE, reported by the Pediatric PSC Consortium. The aim of this study was to create and validate a predictive model for pediatric PSC using data from over 1,000 children. A scoring system was generated based on bilirubin, albumin, platelets, GGT, and presence of large duct disease, and the cohort was risk stratified based on these scores, with zero to three being low-risk group, four to five medium-risk, and six to 11 high-risk group. Now, we found that the SCOPE score positively correlated with the degree of fibrosis on liver histology as shown here. SCOPE scores of zero to four were associated with predominantly F0 to F2 fibrosis stages, while SCOPE scores of six or greater had a much higher frequency of F3 and F4 stage fibrosis. The SCOPE index can accurately predict the probability of the need for transplant or the occurrence of a hepatobiliary complication within a five-year time span, and I think the SCOPE score will be an important tool for families in predicting the risk of a PSC-related complication. I've highlighted in the black boxes an example of how to use this tool. If the child has a SCOPE score of five, then there's a 10% probability of needing a transplant and a 25% probability of having a hepatobiliary complication five years later. Comparison of the SCOPE with other PSC prognostic models revealed that the SCOPE score at diagnosis, as well as two years after diagnosis, was superior to the other adult prognostic models in predicting outcomes for children with PSC. So in summary, the key takeaway points are that up to 40% of children will have clinical progression of disease within 10 years. PSC IBD patients have diminished growth, but less severe IBD than IBD-only patients. Neither UDCA nor vancomycin affected long-term outcomes in retrospective analyses. And both the MRCP scoring system and the SCOPE index are powerful prognostic tools. So where do we go from here? As I've shown you, the vast majority of pediatric PSC studies are retrospective and in order to accurately characterize various aspects of the disease, it's essential to develop prospective studies. To that end, the Childhood Liver Disease Research Network will soon be starting a prospective observational study of PSC in children. The aims of this study will be, one, to characterize the various phenotypes and the impact of these phenotypes on outcomes. Two, to characterize the global functional health of these children. Three, to assess the accuracy of imaging modalities such as fibroscan, MR elastography, and 3D MRCP in predicting outcomes. And four, the creation of a repository of biospecimens for future mechanistic studies. Thank you. Good evening, everyone. It's a great pleasure to be part of this interesting symposium. My task tonight is to talk about immune cholangiopathies when it is not primary sclerosing cholangitis. I have no conflicts of interest, but I have a number of national and international roles relevant to this presentation. So I'm going to focus on immune cholangiopathies, in particular autoimmune sclerosing cholangitis, IgG4-associated cholangitis, post-transplant cholangitis, and touch very briefly on drug-induced cholangiopathies, AIDS, and of course, as a pediatrician, biliary atresia. So there are many different forms of cholangiopathies. They've got a very different etiology, and I'm not going to talk about genetic, infectious, vascular, or malignant cholangiopathies, but in essence, all of these diseases involve progressive destruction of intra- and extubated bile ducts, resulting in a reduction in interlobular bile ducts, or vanishing bile duct disease, fibrosis, cirrhosis, and chronic cholestatic failure. Now, over the years, we've learned a great deal about cholangiocytes. We've always known how important they were for bile production and homeostasis, and we now know that they have an essential role in liver regeneration, and that they're responsible for both epithelial, innate, and adaptive immunity. And in many cases, they are our first line of defense against pathogens using receptors such as toe-like receptors. And as pediatricians looking after children with biliary diseases, we know that reactive ductular cells have a big contribution towards liver repair, inflammation, and fibrosis. Now, I'm going to start with autoimmune liver disease in children, and there's a big spectrum from autoimmune hepatitis to sclerosing cholangitis, whether it's considered autoimmune or not, and overlap syndrome, which is essentially autoimmune hepatitis with biliary features. And this disease has a very close association with inflammatory bile disease. Now, autoimmune liver disease is not that common in children, not as common as in adults. Autoimmune hepatitis is more common than autoimmune sclerosing cholangitis, which is more common than PSC. Now, in this audience is very familiar with the features of sclerosing cholangitis. We know about the clinical features of chronic hepatitis, cirrhosis, portal hypertension, and that it's associated with inflammatory bowel disease. These are all not new. And what differentiates autoimmune liver disease from primary sclerosing cholangitis is essentially the histology, because this very characteristic onion skinning of the bile ducts is actually a very late feature, and there's significant interface hepatitis and plasma cell infiltration. The biliary imaging, which we can see here on this MRI, demonstrates characteristic enlarged gallbladder and these abnormal bile ducts. But what characterizes this as autoimmune is the presence of nonspecific autoantibodies, ANA and SMA, and the plasma cell infiltrate. P-anchor is associated with sclerosing cholangitis with or without inflammatory bowel disease. So biochemically, sclerosing cholangitis is a bit different from hepatitis. There's a biliary picture with elevated alkaline phosphatase and gamma-GT, and synthetic function is preserved until quite late. Treatment of sclerosing cholangitis consists of supportive treatment and ursodeoxycholic acid, and in those children that we consider have autoimmune liver disease, we add prednisolone and azathioprine. The disease doesn't do very well in the long run, so it's important to consider liver transplantation quite early. In King's College some years ago, they did a prospective study comparing autoimmune hepatitis and autoimmune sclerosing cholangitis in children. It's not a large study, but it really has stood the test of time, and you can see that there are similarities, that both groups of children have nonspecific autoantibodies, but there's a higher rate of P-anchor in children with autoimmune sclerosing cholangitis. So are they one disease, or are they two diseases? Well, ESPGAN, the European Pediatric Gastroenterology Society, brought out a position paper two years ago, and they very clearly think that autoimmune sclerosing cholangitis is a separate disease, and they provide a definition for this, which I've discussed with you already, and they differentiate it from sclerosing cholangitis. However, they do recognize its very close association with inflammatory bowel disease, and part of the position statement is to include in the diagnosis cholangiography, fecal calprotectin, and colonoscopy looking for inflammatory bowel disease. Also published in this position paper is a scoring criteria for the diagnosis of juvenile autoimmune liver disease, and also to differentiate between autoimmune hepatitis and autoimmune sclerosing cholangitis. This score really needs to be validated, but it's a useful place to start. Now, are autoimmune sclerosing cholangitis and autoimmune hepatitis the same disease, or are they different diseases? I've shown you that they have very similar clinical, biochemical, and immunological presentations. They have the same immune presentation and the same response to immunosuppression, but are they a different disease, or are they one disease with different outcomes? We reviewed our patients in Birmingham, and we've done a retrospective study. These are very preliminary data today that I'm presenting to you, but we looked at children at three time points, at disease onset, at the time of transition to adult services, and at the last adult clinic follow-up, and we followed up children for between 20 to 28 years. They're classified in the same four groups that I've already mentioned, and we excluded children with any other form of liver disease. We followed now 157 patients with newly diagnosed autoimmune liver disease. The median age at presentation, as you might imagine, is 12 years, and more of them were female than were male. As we know from the literature, the majority had autoimmune liver disease. Far less had autoimmune type 2. Even less had PSC, and between them, overlap hepatitis and ASC were 20% of the children. But what's really interesting is to look at the diagnosis over time. You can see here that this is the characteristic of the four different sorts of liver disease in a bar chart, autoimmune hepatitis 1, 2, overlap, and PSC. You can see that at the time of transition to adult services, and even more strikingly, at the time of their last adult clinic follow-up, that far fewer of the children still had a diagnosis of type 1, and less had a diagnosis of type 2. But you can see there's an increase in children being diagnosed as having ASC and overlap, and in time, far more of them with PSC. When we look at the patient and native liver survival, as we might expect, those with type 2 do worse, are less likely to survive, and are more likely to have a liver transplant. But looking again at PSC and overlap, they tend certainly in early stages to do less well than children with autoimmune hepatitis type 1. So we haven't been the only ones to recognize that autoimmune liver disease progresses in adult life, and this is a study published in Hepatology seven years ago. You can see that those with autoimmune hepatitis are fairly constant, and an increase in those characterized as ASC and PSC. But does this tell us whether they are the same disease or not? Well, we know that ASC and overlap probably are precursors of PSC, and they probably are the same disease. But there are a number of children who very definitely have autoimmune hepatitis who appear to develop PSC in adult life. Whatever the reasons for this, there is a worse outcome for those with ASC, PSC, and overlap, with much higher rates of transplantation in adult life. Moving on now to look at IgG4-related disease, and this is a fairly recently described disease which affects very many different parts of the body. But today we're just focusing on the liver and the bile ducts, and IgG4-related sclerosing cholangitis, or ISC, presents, as you might imagine, with obstructive jaundice involving predominantly the proximal bile ducts. It is a disease of older men that affects men more than women, and there are a large number of reports from Japan. To my knowledge, it's not been reported in children and adolescents, although I look for it every time I see a child with PSC, and it has a very close association with IgG4-related pancreatitis. We know that the etiology is related to altered immunity, but whether there's a genetic susceptibility for this immunity is not very clear. But a number of studies have looked at the immunology, and it's predominantly a Th2-dominant immune reaction with regulatory T-cell activation and expansion of B-cells and regulatory B-cells, which stimulate, in turn, receptor-mediated phagocytosis and a number of signaling pathways. The diagnosis is pretty straightforward because everybody pretty well has a very high IgG4, and the histology of whatever tissue is being affected has this very characteristic lymphoplastic infiltrate and this characteristic storiform fibrosis with obliterative phlebitis, and that affects every organ that is involved. There is, as I've said, an association with IgG4 autoimmune pancreatitis, but thankfully a very good response to steroids, and this is part of the diagnostic criteria. So treatment is with high-dose corticosteroids. There are anecdotal case reports of azathioprine, MMF, and methotrexate in those who are resistant to steroids. Unfortunately, relapse is common, but it does respond to rituximab and other B-cell depletion agents. The prognosis is overall good with the very low risk of liver failure and the development of Bering malignancy, and in adults the main differential diagnosis is PSC or cholangiocosmoma. I want to talk about the cholangiopathy that we see with HIV disease, and although this is due to the cholangiopathy is due to opportunistic infections, predominantly with cryptosporidium and CMV, I believe it does have an immune etiology, and this is because the cholangiopathy does not improve until the immune function is back to normal or pretty well normal with antiretroviral therapy, and only then will these antibiotics be effective for this condition. It's important that we don't forget drug-induced cholangiopathy. This can also cause immune-mediated damage to interlobular ducts, very similar to primary biliary cholangitis with a progression to biliary cirrhosis. And top of the list for causing these are antibiotics, amoxicillin, pepricillin, imipenem. Carbamazepine has also been reported to cause this syndrome, and the new cancer treatment immune checkpoint inhibitors have also been associated with a drug-induced cholangiopathy. Treatment is to discontinue the medication, but in some cases steroid therapy is required with ursodeoxycholic acid and the monoclonal antibody, Tuclizumab. Those of us who look after transplant patients know that post-transplant cholangiopathies is a big, big issue. I'm not going to mention the work of the biliary complications associated with technical issues or ischemia, but as we all know, there's a very wide range of immune-mediated cholangiopathies associated with acute or chronic rejection or GBHD, the vanishing bile duct syndrome, whether it's stimulated by CMV or rejection, and a more recently described cholangiopathy, antedonal IgG3 and 4 cholangiopathy, which is associated with increased IgG4 subclasses. And all of these resolve with increased immunosuppression or, if not, require re-transplantation. So finally, I must mention bariatresia. This is a very classic, as we know, obstructive cholangiopathy of unknown etiology. And it's not clear whether genetic susceptibility plays a role in this, but we do know that immune genes are upregulated in the condition, this condition with a number of immune genes being overexpressed. And we know from animal models that it may be associated with viral infections. So it's possible that biliatresia is an immune cholangiopathy with abnormal adaptive immune response leading to bile duct injury. So thank you, ladies and gentlemen, and friends and colleagues. I've summarized for you a very fascinating field of immune cholangiopathies, diseases of which we're just now beginning to understand the varied etiology and pathogenesis and put our knowledge of cholangiocytes and their role in immunity and regeneration at the very focus of this cholangiopathy. And I hope that all that we have learned so far will improve our understanding outcome and identify potential targets for therapy and improve outcomes for our patients. Thank you very much. Hello. Thank you for attending the Cholestatic and Autoimmune Liver Disease SIG program during the liver meeting 2020. I'd like to thank the AASLD and the course organizers, Dr. Shah and Milo, for inviting me to present today. My presentation is entitled Pharmacological Management, What is in the Pipeline? My name is Cynthia Levy. I'm professor of clinical medicine at the University of Miami. And these are my disclosures. There are multiple opportunities for disease modulation in PSC, with several different targets currently being explored, as shown here. And within each of these classes of drugs, there are several compounds under evaluation, some with multiple mechanisms of action. Unfortunately, we can't discuss all of these drugs today, so I chose to focus on those undergoing phases 2 to 3 testing. This slide shows the current pipeline in PSC. Leading the board, we have norursol, salofaxer, and vancomycin entering phase 3 trials. sulfasalazine, the hightined compound, the pliant compound 74809, and an iBAT inhibitor entering phase 2 trials. Let's start with norUDCA. Understanding how different it is from UDCA and familiarizing ourselves with its mechanisms of action. NorUDCA is a side chain shortened homolog of UDCA, and this modification renders the drug resistant to conjugation with taurine and glycine. Thus, the unconjugated molecule can be absorbed through the cholangiocyte into sinusoids and back into the hepatocyte, creating what is called cholepatic shunting. This process results in ductal targeting and generates bio that is enriched with bicarbonate, which is cytoprotective. In animal models, norUDCA had profound anti-cholestatic, anti-inflammatory, and anti-fibrotic properties, more so than UDCA. In addition, recent studies revealed an additional mechanism of action for norUDCA through modulation of mTOR proteome to redirect metabolic sensing programs. This way, norUDCA can suppress effector functions of CD8 and shift CD4 cells to counteract Th17-Treg balance. This supports the strong immunomodulatory effect of norUDCA. In a phase 2 trial conducted in Europe, 161 patients were randomized to placebo versus one of three doses of norUDCA and treated for 12 weeks. A reduction in alkaline phosphatase, NGGT, was observed in a dose-dependent fashion, reaching 26% reduction with the highest dose of 1,500 mg daily. Importantly, this effect was indicated by the fact that the dose-dependent importantly, this effect was independent to prior exposure or response to UDCA, and the safety profile was similar to placebo. No itching was observed. Based on these results, a phase 3 study was launched utilizing the 1,500 mg daily dose, and 300 patients are expected to be treated for 96 weeks. Endpoints will be biochemical, histological, and clinical. Moving on to FXR agonists, activation of this nuclear receptor modulates various signaling pathways. Importantly, it affects bioacid homeostasis primarily by inhibiting bioacid synthesis and regulating genes involved in the bioacid metabolism, creating a choleretic effect. We have increased in expression of transporters involved in canalicular and basolateral bioacidic flux, and have reduced expression of transporters involved in basolateral bioacid uptake. Anti-inflammatory and anti-fibrotic effects are also described. Obeticholic acid is the first in class FXR agonist. A 24-week phase 2 randomized control trial evaluating two different doses of OCA versus placebo demonstrated a significant reduction in serum alkaline phosphatase, which was more remarkable with the highest dose of 5 to 10 mg daily. Fluoride was observed in 67% of patients receiving this dose compared to 46% of patients on placebo. Despite some evidence for effectiveness, to my knowledge, this drug is not moving forward in the evaluation process, but a similar compound is. Silofaxor is a second generation FXR agonist, which was evaluated in a small phase 2 trial for non-serotic patients with PSC, and who also had elevated alkaline phosphatase levels. 52 patients were randomized to two different doses of silofaxor versus placebo in a 2 to 2 to 1 fashion and treated for 12 weeks. Indeed, we observe a reduction in alkaline phosphatase from baseline, which was more pronounced with the highest dose of 100 mg daily. This effect was also independent of UDCA use. Fluoride was observed in 36% of patients treated with 100 mg daily dose versus 60% of placebo treated patients. Based on this promising result, silofaxor is now undergoing further evaluation on a large phase 3 trial with planned enrollment of 400 patients using the dose of 100 mg daily for 96 weeks. Primary endpoint will be histological. And there's also a phase 1 trial with a plan to enroll 20 patients to evaluate safety and tolerability of various doses of silofaxor for 12 weeks in patients with compensated cirrhosis. So here the primary outcome will be basically looking at treatment emerging adverse events, as well as laboratory abnormalities. Moving on to vancomycin. The background information for use of vancomycin in PSC comes from knowledge that gut microbiota has been implicated in the pathogenesis of PSC. Patients with PSC have decreased diversity in gut microbiota compared to healthy controls and also to patients with IBD alone. And vancomycin is an antimicrobial agent that also has a monomodulatory effect. It reduces cytokine release from T cells and increases the number of Tregs. Preliminary studies with vancomycin in adults show a significant reduction in alkaline phosphatase, noting better response and greater tolerability with the low dose, which is 125 milligrams, four times a day. With that dose, there is also improvement in the male risk score. However, these studies were small and of short duration of about 12 weeks. In children, case series also suggests that vancomycin leads to higher rates of GGT and ALT normalization, in addition to an increasing peripheral pose of Treg cells. However, we must now also take into consideration the controversial data from the Pediatric PSC Consortium, where 88 patients treated with vancomycin were propensity matched to patients treated with EDCA or who were on observation only. Those were analyzed by intention to treat after one year of therapy. GGT normalized in 53% of vanco-treated patients, 49% of EDCA-treated patients, and 52% of untreated patients. In addition, the five-year probability of liver transplant listing was not statistically different between groups, even if you focus only on the high-risk patients. The graphs below show changes in GGT, total bilirubin, and APRI. You can see improvement of GGT at six months and one year in patients treated with vancomycin, but also in patients treated with EDCA and in observation. There was no significant difference between groups. These controversial results highlight the importance of a well-conducted randomized control trial. Indeed, we now have a large randomized control trial currently enrolling patients at the three male clinics. The target recruitment is 102 patients with large duct PSC and high alkaline phosphatase. The primary endpoint will be alkaline phosphatase normalization after 18 months of treatment. Let's now move on to the drugs undergoing phase 2 testing, starting with the Hytide-1801, which is a combination of berberine and EDCA. Berberine is a traditional Chinese medication with proposed metabolic and antimicrobial effects. In addition, berberine seems to potentiate EDCA effects. In animal models of bioduct ligation, this compound consistently ameliorates hepatocellular necrosis, portal inflammation, and bioduct proliferation. As you can see in the bar graphs below, animals treated with the compound had reduced levels of alkaline phosphatase, increased levels of superoxide dismutase, thus better antioxidant activity, and reduced MDA levels, thus less oxidative stress. A proof-of-concept study was just completed evaluating 90 patients who were treated with Hytide compound 500 milligrams twice a day or 1,000 milligrams twice a day compared to placebo for a total of 18 weeks. We are anxiously waiting for the results to be reviewed. Sulfasalazine is a well-known anti-inflammatory drug, which is also undergoing evaluation for PSC. This study plans to enroll 42 patients in a 14-week randomized controlled trial, followed by six weeks of open-label treatment, with the primary endpoint being biochemical. It is unclear exactly how sulfasalazine exerts its anti-inflammatory properties, but the proposed mechanisms of action in PSC include free radical scavenging, impairment of anti-TNF binding to its receptor, reduction in B-cell antibody production, and possibly blocking the bacterial enzymes that produce hydrogen sulfide. We know that about 10% of sulfasalazine enter the systemic circulation unaltered and therefore can reach the liver, where it could exert these effects. There is also anecdotal data and observations from small case control studies like this one presented during DDW this year, where patients treated with sulfasalazine, patients with PSC and IBD treated with sulfasalazine were compared to those treated with mesalamine, if they had elevated alkaline phosphatase at baseline. And the authors observed a higher proportion of patients treated with sulfasalazine achieving biochemical response compared to patients treated with mesalamine. Switching gears to a completely innovative and exciting target, the plant compound is an anti-integrin that specifically targets alpha-B-beta-6 and alpha-B-beta-1 integrins, which are particularly expressed in the cholangiocyte and fibroblasts. These integrins promote fibrosis through activation of TGF-beta, which will then lead to transcription of a number of pro-fibrotic genes. So blocking these integrins is expected to profoundly inhibit fibrogenesis through reduction of expression of multiple pro-fibrotic genes. And finally, I'd like to direct you to the ASBT inhibitors. We're not discussing these drugs as disease-modifying agents, but as treatment for one of the most distressing symptoms these patients have, namely itching. The ASBT inhibitors inhibit enterohepatic circulation of bio-acids, and preliminary studies suggest they can reduce serum levels of bio-acids and pruritus, as was the case for this phase 2 proof-of-concept study with maralexibat. Focusing on the graph on the left, one can observe a 70% reduction in pruritus among patients whose itch score was 4 or higher at baseline. On the right, we can see that there was also a significant reduction on the levels of serum bio-acids. With that in mind, a phase 2 randomized control trial is planned to study efficacy and safety of another ASBT inhibitor called volixibat in the treatment of cholestatic itching. The anticipated study design is shared here, and attention to four months of blinded treatment followed by an open-label extension that is planned to run for two years. Patients with small duct PSC and compensated sebroses will be allowed in the study, as well as concomitant IBD and use of UDCA. So it will be an interesting study. In conclusion, this is an exciting time for patients with PSC, with various existing targets and mechanisms of action being explored. There are several drugs undergoing phase 2 to 3 testing, and many of these designs also provide an opportunity for us to better understand the disease pathogenesis and natural history. I am sure there is a lot to learn in the coming years, hopefully leading to a new drug approval soon. Thank you for your attention. Hello. My name is John Eaton. I'm a hepatologist for Mayo Clinic in Rochester, Minnesota. I'd like to thank the moderators for inviting me to speak with you about novel imaging in primary sclerosine cholangitis, or PSC, both today and tomorrow. Our objectives are to highlight the current clinical application and the role of magnetic resonance imaging in the management of our patients with PSC as it stands today, and review novel approaches as it relates to imaging on the horizon. MRI and MRCP play a central role in the diagnosis and management of those with PSC. It is the diagnostic test of choice for PSC and can readily detect benign complications associated with this condition. Moreover, imaging can play a useful tool in triaging individuals towards ERCP and potentially avoiding an ERCP in some instances, and serve as a roadmap for an endoscopist, highlighting areas which should be targeted for therapeutic interventions or sampling, and also in some instances highlighting areas which may be best to avoid during the ERCP. And lastly, MRI and MRCP can detect hepatobiliary malignancies associated with PSC. This is illustrated on the axial MRI images shown here. The image on the right is a delayed phase image, and you can see on the delayed phase image progressive enhancement of a perihylar mass lesion, which is characteristic of a perihylar clandiocarcinoma, the leading source of mortality among those with PSC. Compared to ultrasound, MRCP is superior for the early detection of perihylar clandiocarcinoma among those with PSC. This improved diagnostic performance also translates into improved outcomes. Detection of perihylar clandiocarcinoma prior to the onset of symptoms utilizing MRCP is associated with improved outcomes, including improvement in overall patient survival following a diagnosis perihylar clandiocarcinoma. These prognostic benefits persisted in a subgroup of patients who had had PSC for greater than one year in duration and who were already enrolled in a regular surveillance program. Magnetic Resonance Elastography, or MRE, is a quantitative MRI technology which measures liver stiffness. Liver stiffness is a surrogate for liver fibrosis. Other sources of an increased liver stiffness can include an acute biliary obstruction, severe hepatic inflammation, or heart failure. MRE utilizes an active driver, which delivers acoustic waves to a passive driver resting on a patient inside a standard MRI scanner. Vibrations travel through the passive driver and into the patient, and the speed in which those vibration waves travel through tissue is illustrated here on this displacement map. The faster those waves travel through tissue, the stiffer that tissue is, and that is then illustrated here on a stiffness map, where stiffer areas are highlighted by brighter colors such as orange or red, and areas of decreased stiffness are highlighted by the cooler colors. Compared to transient elastography, MRE has been shown to have a reduced failure rate and importantly, samples a significantly larger volume of the liver, which may be important in a notoriously patchy disease such as PSC. In addition, MRE has the potential to detect other PSC-associated complications during the same time, and can also be performed along with an MRCP without adding a significant amount of time to the MRI exam, and often at minimal or no added costs. Liver stiffness measured by MRE is associated with various stages of fibrosis and PSC, but it's also associated with outcomes. In fact, both the baseline liver stiffness, greater than 4.3 kilopascals, and changes in liver stiffness over time are highly predictive of the development of hepatic decompensation, with a C-statistic of 93%. Now that we've discussed the role of imaging and day-to-day management of those with PSC today, let's talk about the role of imaging and novel technologies of the future. In contrast to the more ubiquitous 2D MRE, which we've discussed previously, 3D MRE has the advantage where it can visualize the liver in its entirety. Moreover, it can examine additional biomechanical properties beyond simply just the liver stiffness, and give insight into the relative contributions of inflammation versus hepatic fibrosis, and how they may be contributing to the overall liver stiffness. Compared to 2D MRE, it may be more accurate in the diagnosis of advanced fibrosis, at least in NAFLD. To date, this has not been studied in patients with PSC, but remains a promising technology of the future. The hepatic-parenchymal enhancement pattern, with the use of hepatocyte-specific contrast agents in the hepatobiliary phase, may provide insight into disease severity and serve as an indirect measure of liver function. The individual on the left has relatively mild PSC, and there is relative uniform enhancement of the hepatic parenchyma in the hepatobiliary phase with a hepatocyte-specific contrast agent. Contrast that to the individual on the right, who has more advanced PSC, with areas of atrophy. There is more patchy enhancement throughout the liver, where areas which are hypoenhancing may correspond to atrophic segments of the liver. Indeed, a diminished relative enhancement after contrast injection may be associated with adverse short-term clinical outcomes. It's likely that the appearance of not only just the liver, but the bile ducts may be important. This was illustrated by a study that looked at a qualitative MRCP prognostic score, called the ANALI score, in addition to elastography measured by transient elastography. The components of the ANALI score include the degree of intrapatic bile duct dilation, hepatic-parenchymal features, and the presence of portal hypertension. Individuals with a low ANALI score and low liver stiffness were less likely to develop adverse outcomes when compared to those individuals with a low ANALI score and high liver stiffness, or vice versa, and compared to those who had a high ANALI score and high liver stiffness. This indirectly suggests that having a composite MRI biomarker examining both the liver and the bile ducts and stiffness may have merit and may be important. The biliary metrics have the potential to provide detailed quantitative measurements of the biliary tree. MRCP Plus is a proprietary technology which does just that. It is able to provide detailed assessments of the biliary tree volume, stricture measurements, and ductal dilation measurements, among many other details. To date, this has been studied largely in small samples of patients with PSC, but this may be able to measure changes over time, as illustrated here, where the degree of strictures and dilations and their abundance changed over time when measured serially in a small cohort of patients with PSC. Indeed, in a small pilot study, MRCP Plus-derived measurements which examined the degree of ductal dilations was associated with outcomes where individuals who had a high burden of ductal dilations were more likely to suffer from adverse events. This promising technology warrants further study in well-designed larger samples. Let's discuss approaches to analyzing imaging data. It can be said that if a picture is worth a thousand words, it can also be said that a picture is worth many millions of bits of data. Magnifying glass is simply a process of extracting quantifiable data from images. In effect, it's like a magnifying glass that uncovers quantifiable data that's otherwise unseen to the human eye. It's simply a process which requires a large number of patients and images which are processed for consistency and quality, and then regions of interest within those images are selected. The regions of interest may be selected through a manual process or a software-derived automated process. And then, within those regions of interest, features are extracted, and these features can include things that we think about as clinicians, such as a size or location. But importantly, it may also include things that we don't tend to think about, such as computational features. This would include things such as the intensity or texture statistics and the relationship between voxels, things that are otherwise unseen to the human eye. Those features are then analyzed using artificial intelligence or traditional statistical methods to help better predict an outcome of interest or enhance the diagnostic performance. Indeed, radiomics can give insight into pathobiology and outcomes, and here's one example. This highlights the use of an MRI radiomic approach, which was able to detect, just based on imaging features alone, the molecular subtype of a glioblastoma with a high degree of accuracy, and similarly was able to risk stratify patients with glioblastoma and predict long-term outcomes. This approach does have promise in its application in PSC and other liver diseases. Artificial intelligence is becoming increasingly ubiquitous in our everyday lives. Simply put, AI is any technique that enables a computer to process, interpret, and learn from data. Machine learning is a subset of AI, which has algorithms that have an ability to learn without being programmed. Deep learning is a subset of machine learning, where algorithms permit software to train itself and learn from vast amounts of data, including graphical data, using neural networks. Deep learning is better suited at analyzing even larger amounts of data compared to traditional machine learning techniques, and tends to lend itself for processing graphical data. Here's a simplified example of deep learning in action. Let's say that we wanted to train a deep learning algorithm to recognize a dog from other animals. You would start the process by feeding an algorithm many hundreds of thousands of images of a dog and other animals. The algorithm would then extract features from these images and filter it through many thousands of layers or deep neural networks. Errors would occur during this process, and the algorithm would learn from these errors and during the subsequent reiteration would become increasingly more accurate. This process would be repeated many times. Deep learning has been applied in the clinical setting. One example is where a deep learning algorithm was created by analyzing many thousands of dermatologic images, and this deep learning algorithm was able to accurately predict the presence of malignant skin lesions and actually outperform the average community dermatologist. Artificial intelligence has also been applied in the study of PSC. Indeed, there was a validated machine learning prediction score called PRESTO, which was created using a machine learning approach, and it utilized routine laboratory variables and clinical variables and was highly accurate in its ability to predict hepatic decompensation. Also, a fully automated deep learning algorithm was trained and validated to diagnose PSC based on MRCP images with a high degree of sensitivity and specificity. In the future, it's possible that big data, including imaging data, can be analyzed using artificial intelligence methods to pave the way for new discoveries in clinical applications, including the discovery of new biomarkers, which may play a role in the conduct of clinical trials, and also pave the way for new translational discoveries. Combining clinical data with imaging and analyzing this with artificial intelligence may have real-world clinical applications in the future. Imagine you're seeing a patient with PSC in the clinic for worsening paritis and fatigue who undergoes an MRCP and MR elastography. The patient's clinical data, laboratory data, imaging data, is automatically uploaded into a validated deep learning algorithm, which then provides a predictive output and a decision support tool, stating that the one-year probability of cholangiocarcinoma is 85%, and it recommends proceeding with an ERCP with biliary brushings and considering a transplant referral. Several key takeaways. Imaging and PSC today centers around the use of MRI and MRCP for the detection of PSC and related complications. The diagnosis of cholangiocarcinoma, while patients are still asymptomatic, utilizing MRI and MRCP has improved outcomes. Elastography is a quantitative MRI technology, which provides relevant prognostic information for our patients. Imaging and PSC tomorrow will continue to center around the use of advancing quantitative MRI technologies, and it's likely that imaging will increasingly be used to risk stratify patients and serve as a surrogate endpoint in clinical trials. Imaging being analyzed with AI and radiomic approaches may pave the way for new biomarker and translational discoveries, and also create new and automated clinical applications and decision support tools that will ultimately enhance the quality of care for our patients with PSC. On behalf of the moderators and my fellow presenters, we'd like to thank you for your attention and for joining us, and we'd like to remind you that these presentations will remain available for the next 90 days after the meeting on demand. Thank you.

Cholestasis

Autoimmune liver diseases

Primary sclerosing cholangitis

Fecal microbiota transplantation

Gut microbiome manipulation

Cholangiocarcinoma

Pediatric PSC

Autoimmune cholangiopathies

Immune cholangiopathies

Cholangiocytes

Liver immunity

Ursodeoxycholic acid