Improving predictive accuracy in primary biliary cholangitis: A new genetic risk score.

BACKGROUND AND AIMS: Genetic variants influence primary biliary cholangitis (PBC) risk. We established and tested an accurate polygenic risk score (PRS) using these variants. METHODS: Data from two Italian cohorts (OldIT 444 cases, 901 controls; NewIT 255 cases, 579 controls) were analysed. The latest international genome-wide meta-analysis provided effect size estimates. The PRS, together with human leukocyte antigen (HLA) status and sex, was included in an integrated risk model. RESULTS: Starting from 46 non-HLA genes, 22 variants were selected. PBC patients in the OldIT cohort showed a higher risk score than controls: -.014 (interquartile range, IQR, -.023, .005) versus -.022 (IQR -.030, -.013) (p < 2.2 × 10-16). For genetic-based prediction, the area under the curve (AUC) was .72; adding sex increased the AUC to .82. Validation in the NewIT cohort confirmed the model's accuracy (.71 without sex, .81 with sex). Individuals in the top group, representing the highest 25%, had a PBC risk approximately 14 times higher than that of the reference group (lowest 25%; p < 10-6). CONCLUSION: The combination of sex and a novel PRS accurately discriminated between PBC cases and controls. The model identified a subset of individuals at increased risk of PBC who might benefit from tailored monitoring.

Adequate versus deep response to ursodeoxycholic acid in primary biliary cholangitis: To what extent and under what conditions is normal alkaline phosphatase level associated with complication-free survival gain?

BACKGROUND AND AIMS: Normal alkaline phosphatase (ALP) levels in ursodeoxycholic acid (UDCA)-treated patients with primary biliary cholangitis (PBC) are associated with better long-term outcome. However, second-line therapies are currently recommended only when ALP levels remain above 1.5 times the upper limit of normal (×ULN) after 12-month UDCA. We assessed whether, in patients considered good responders to UDCA, normal ALP levels were associated with significant survival gains. APPROACH AND RESULTS: We performed a retrospective cohort study of 1047 patients with PBC who attained an adequate response to UDCA according to Paris-2 criteria. Time to liver-related complications, liver transplantation, or death was assessed using adjusted restricted mean survival time (RMST) analysis. The overall incidence rate of events was 17.0 (95% CI: 13.7-21.1) per 1000 out of 4763.2 patient-years. On the whole population, normal serum ALP values (but not normal gamma-glutamyl transpeptidase (GGT), alanine aminotransferase (ALT), or aspartate aminotransferase (AST); or total bilirubin < 0.6 ×ULN) were associated with a significant absolute complication-free survival gain at 10 years (mean 7.6 months, 95% CI: 2.7 - 12.6 mo.; p = 0.003). In subgroup analysis, this association was significant in patients with a liver stiffness measurement ≥ 10 kPa and/or age ≤ 62 years, with a 10-year absolute complication-free survival gain of 52.8 months (95% CI: 45.7-59.9, p < 0.001) when these 2 conditions were met. CONCLUSIONS: PBC patients with an adequate response to UDCA and persistent ALP elevation between 1.1 and 1.5 ×ULN, particularly those with advanced fibrosis and/or who are sufficiently young, remain at risk of poor outcome. Further therapeutic efforts should be considered for these patients.

A quantitative MRCP-derived score for medium-term outcome prediction in primary sclerosing cholangitis.

BACKGROUND: Magnetic resonance cholangiopancreatography (MRCP) is the gold standard for diagnosis of patients with primary sclerosing cholangitis (PSC). The semi-quantitative MRCP-derived Anali scores proposed for risk stratification, have poor-to-moderate inter-reader agreement. AIMS: To evaluate the prognostic performance of quantitative MRCP metrics in PSC. METHODS: This is a retrospective study of PSC patients undergoing MRCP. Images were processed using MRCP+ software (Perspectum Ltd, Oxford) that provides quantitative biliary features, semi-automatically extracted by artificial intelligence-driven analysis of MRCP-3D images. The prognostic value of biliary features has been assessed for all hepato-biliary complications. RESULTS: 87 PSC patients have been included in the analysis. Median follow-up from MRCP to event/censoring of 30.9 months (Q1-Q3=13.6-46.6). An adverse outcome occurred in 27 (31.0%) patients. The number of biliary strictures (HR=1.05 per unit, 95%CI 1.02-1.08, p < 0.0001), spleen length (HR=1.16 per cm, 95%CI 1.01-1.34, p = 0.039), adjusted for height, age at MRCP, and time from diagnosis to MRCP predicted higher risk of hepatobiliary complications. These were incorporated into a the quantitative MRCP-derived PSC (qMRCP-PSC) score (C-statistic=0.80). After 3-fold cross-validation, qMRCP-PSC outperformed the Anali score in our cohort (C-statistic of 0.78 vs 0.64) and enabled the discrimination of survival of PSC patients (log-rank p < 0.0001). CONCLUSIONS: The qMRCP-PSC score identified patients at higher risk of hepatobiliary complications and outperformed the available radiological scores. It represents a novel quantitative biomarker for disease monitoring and a potential surrogate endpoint for clinical trials.

Risk stratification in primary sclerosing cholangitis.

Primary sclerosing cholangitis (PSC) is a chronic liver disorder commonly affecting young patients and associated with uncertain prognosis and elevated risk of end-stage liver disease and hepatobiliary cancer. Rate of progression in PSC is heterogeneous and accurately predicting the disease course is of paramount importance to clinical practice and interventional trial design. So far, efforts have brought to the development of models looking at short-to-middle-term outcome using composite models including clinical, laboratory, radiological and histological parameters with limited performance. In the era of whole genome sequencing and digital innovation, the time is ripe for the development of stratified medicine in PSC. Efforts should be directed toward developing well-phenotyped cohorts of patients with longitudinal follow-up across sustained periods of time, application of novel image-processing technology, and biomarker discovery using multiomics platforms.

European Society for Organ Transplantation (ESOT) Consensus Statement on Outcome Measures in Liver Transplantation According to Value-Based Health Care.

Liver transplantation is a highly complex, life-saving, treatment for many patients with advanced liver disease. Liver transplantation requires multidisciplinary teams, system-wide adaptations and significant investment, as well as being an expensive treatment. Several metrics have been proposed to monitor processes and outcomes, however these lack patient focus and do not capture all aspects of the process. Most of the reported outcomes do not capture those outcomes that matter to the patients. Adopting the principles of Value-Based Health Care (VBHC), may provide an opportunity to develop those metrics that matter to patients. In this article, we present a Consensus Statement on Outcome Measures in Liver Transplantation following the principles of VBHC, developed by a dedicated panel of experts under the auspices of the European Society of Organ Transplantation (ESOT) Guidelines' Taskforce. The overarching goal is to provide a framework to facilitate the development of outcome measures as an initial step to apply the VMC paradigm to liver transplantation.

Artificial intelligence for precision medicine in autoimmune liver disease.

Autoimmune liver diseases (AiLDs) are rare autoimmune conditions of the liver and the biliary tree with unknown etiology and limited treatment options. AiLDs are inherently characterized by a high degree of complexity, which poses great challenges in understanding their etiopathogenesis, developing novel biomarkers and risk-stratification tools, and, eventually, generating new drugs. Artificial intelligence (AI) is considered one of the best candidates to support researchers and clinicians in making sense of biological complexity. In this review, we offer a primer on AI and machine learning for clinicians, and discuss recent available literature on its applications in medicine and more specifically how it can help to tackle major unmet needs in AiLDs.

Role of ductular reaction and ductular-canalicular junctions in identifying severe primary biliary cholangitis.

Background & Aims: Primary biliary cholangitis (PBC) is a chronic cholangiopathy characterised by immuno-mediated injury of interlobular bile ducts leading to intrahepatic cholestasis and progressive liver fibrosis. PBC histology is characterised by portal inflammation, progressive fibrosis, ductopenia, and the appearance of the so-called ductular reaction. The aim of the present study was to investigate the pathogenetic relevance of ductular reaction in PBC. Methods: Liver biopsies were collected from naïve people with PBC (N = 87). Clinical-serological parameters were obtained at diagnosis and after 1 year of ursodeoxycholic acid (UDCA) treatment. Histological staging was performed on all slides according to multiple scoring systems and criteria for PBC. Liver samples were obtained from Mdr2 -/- mice treated with or without UDCA. Samples were processed for histology, immunohistochemistry, and immunofluorescence. Results: Ductular reaction in people with PBC correlated with the disease stage and liver fibrosis, but not with disease activity; an extensive ductular reaction correlated with serum alkaline phosphatase levels at diagnosis, response to UDCA, and individuals' estimated survival, independently from other histological parameters, including disease stage. In people with PBC, reactive ductules were associated with the establishment of junctions with bile canaliculi and with fibrogenetic cell activation. Consistently, in a mouse model of intrahepatic cholestasis, UDCA treatment was effective in reducing ductular reaction and fibrosis and increasing ductular-canalicular junctions. Conclusions: Extensive ductular reaction outlines a severe histologic phenotype in PBC and is associated with an inadequate therapy response and a worse estimated prognosis. Lay summary: In people affected by primary biliary cholangitis (PBC), the histological appearance of extensive ductular reaction identifies individuals at risk of progressive fibrosis. Ductular reaction at diagnosis correlates with the lack of response to first-line therapy with ursodeoxycholic acid and serves to restore ductular-canalicular junctions in people with PBC. Assessing ductular reaction extension at diagnosis may add valuable information for clinicians.

LLM-PBC: Logic Learning Machine-Based Explainable Rules Accurately Stratify the Genetic Risk of Primary Biliary Cholangitis.

BACKGROUND: The application of Machine Learning (ML) to genetic individual-level data represents a foreseeable advancement for the field, which is still in its infancy. Here, we aimed to evaluate the feasibility and accuracy of an ML-based model for disease risk prediction applied to Primary Biliary Cholangitis (PBC). METHODS: Genome-wide significant variants identified in subjects of European ancestry in the recently released second international meta-analysis of GWAS in PBC were used as input data. Quality-checked, individual genomic data from two Italian cohorts were used. The ML included the following steps: import of genotype and phenotype data, genetic variant selection, supervised classification of PBC by genotype, generation of "if-then" rules for disease prediction by logic learning machine (LLM), and model validation in a different cohort. RESULTS: The training cohort included 1345 individuals: 444 were PBC cases and 901 were healthy controls. After pre-processing, 41,899 variants entered the analysis. Several configurations of parameters related to feature selection were simulated. The best LLM model reached an Accuracy of 71.7%, a Matthews correlation coefficient of 0.29, a Youden's value of 0.21, a Sensitivity of 0.28, a Specificity of 0.93, a Positive Predictive Value of 0.66, and a Negative Predictive Value of 0.72. Thirty-eight rules were generated. The rule with the highest covering (19.14) included the following genes: RIN3, KANSL1, TIMMDC1, TNPO3. The validation cohort included 834 individuals: 255 cases and 579 controls. By applying the ruleset derived in the training cohort, the Area under the Curve of the model was 0.73. CONCLUSIONS: This study represents the first illustration of an ML model applied to common variants associated with PBC. Our approach is computationally feasible, leverages individual-level data to generate intelligible rules, and can be used for disease prediction in at-risk individuals.

SARS-CoV-2 vaccination and risk of severe COVID-19 outcomes in patients with autoimmune hepatitis.

BACKGROUND: Data regarding outcome of Coronavirus disease 2019 (COVID-19) in vaccinated patients with autoimmune hepatitis (AIH) are lacking. We evaluated the outcome of COVID-19 in AIH patients who received at least one dose of Pfizer- BioNTech (BNT162b2), Moderna (mRNA-1273) or AstraZeneca (ChAdOx1-S) vaccine. PATIENTS AND METHODS: We performed a retrospective study on AIH patients with COVID-19. The outcomes of AIH patients who had acute respiratory syndrome coronavirus 2 (SARS-CoV-2) breakthrough infection after at least one dose of COVID-19 vaccine were compared to unvaccinated patients with AIH. COVID-19 outcome was classified according to clinical state during the disease course as: (i) no hospitalization, (ii) hospitalization without oxygen supplementation, (iii) hospitalization with oxygen supplementation by nasal cannula or mask, (iv) intensive care unit (ICU) admission with non-invasive mechanical ventilation, (v) ICU admission with invasive mechanical ventilation or (vi) death, and data was analyzed using ordinal logistic regression. RESULTS: We included 413 (258 unvaccinated and 155 vaccinated) patients (81%, female) with a median age of 52 (range: 17-85) years at COVID-19 diagnosis. The rates of hospitalization were (36.4% vs. 14.2%), need for any supplemental oxygen (29.5% vs. 9%) and mortality (7% vs. 0.6%) in unvaccinated and vaccinated AIH patients with COVID-19. Having received at least one dose of SARS-CoV-2 vaccine was associated with a significantly lower risk of worse COVID-19 severity, after adjusting for age, sex, comorbidities and presence of cirrhosis (adjusted odds ratio [aOR] 0.18, 95% confidence interval [CI], 0.10-0.31). Overall, vaccination against SARS-CoV-2 was associated with a significantly lower risk of mortality from COVID-19 (aOR 0.20, 95% CI 0.11-0.35). CONCLUSIONS: SARS-CoV-2 vaccination significantly reduced the risk of COVID-19 severity and mortality in patients with AIH.

Predictors of serious adverse events and non-response in cirrhotic patients with primary biliary cholangitis treated with obeticholic acid.

BACKGROUND & AIMS: Obeticholic acid (OCA) has recently been restricted in patients with primary biliary cholangitis (PBC) with "advanced cirrhosis" because of its narrow therapeutic index. We aimed to better define the predicting factors of hepatic serious adverse events (SAEs) and non-response in cirrhotic patients undergoing OCA therapy. METHODS: Safety and efficacy of treatment were evaluated in a cohort of consecutive PBC cirrhotic patients started with OCA. OCA response was evaluated according to the Poise criteria. Risk factors for hepatic SAEs and non-response were reported as risk ratios (RR) with 95% confidence intervals (CIs). RESULTS: One hundred PBC cirrhotics were included, 97 Child-Pugh class A and 3 class B. Thirty-one had oesophageal varices and 5 had a history of ascites. Thirty-three per cent and 32% of patients achieved a biochemical response at 6 and 12 months respectively. Male sex (adjusted-RR 1.75, 95%CI 1.42-2.12), INR (1.37, 1.00-1.87), Child-Pugh score (1.79, 1.28-2.50), MELD (1.17, 1.04-1.30) and bilirubin (1.83, 1.11-3.01) were independently associated with non-response to OCA. Twenty-two patients discontinued OCA within 12 months: 10 for pruritus, 9 for hepatic SAEs (5 for jaundice and/or ascitic decompensation; 4 for upper digestive bleeding). INR (adjusted-RR 1.91, 95%CI 1.10-3.36), lower albumin levels (0.18, 0.06-0.51), Child-Pugh score (2.43, 1.50-4.04), history of ascites (3.5, 1.85-6.5) and bilirubin (1.30, 1.05-1.56), were associated with hepatic SAEs. A total bilirubin≥1.4 mg/dl at baseline was the most accurate biochemical predictor of hepatic SAEs under OCA. CONCLUSIONS: An accurate baseline assessment is crucial to select cirrhotic patients who can benefit from OCA. Although OCA is effective in one third of cirrhotics, bilirubin level ≥1.4 mg/dl should discourage from its use.

Liver stiffness measurement by vibration-controlled transient elastography improves outcome prediction in primary biliary cholangitis.

BACKGROUND & AIMS: Liver stiffness measurement (LSM) by vibration-controlled transient elastography (VCTE) has been shown to predict outcomes of patients with primary biliary cholangitis (PBC) in small-size studies. We aimed to validate the prognostic value of LSM in a large cohort study. METHODS: We performed an international, multicentre, retrospective follow-up study of 3,985 patients with PBC seen at 23 centres in 12 countries. Eligibility criteria included at least 1 reliable LSM by VCTE and a follow-up ≥ 1 year. Independent derivation (n = 2,740) and validation (n = 568) cohorts were built. The primary endpoint was time to poor clinical outcomes defined as liver-related complications, liver transplantation, or death. Hazard ratios (HRs) with CIs were determined using a time-dependent multivariable Cox regression analysis. RESULTS: LSM was independently associated with poor clinical outcomes in the derivation (5,324 LSMs, mean follow-up 5.0 ± 3.1 years) and validation (1,470 LSMs, mean follow-up 5.0 ± 2.8 years) cohorts: adjusted HRs (95% CI) per additional kPa were 1.040 (1.026-1.054) and 1.042 (1.029-1.056), respectively (p <0.0001 for both). Adjusted C-statistics (95% CI) at baseline were 0.83 (0.79-0.87) and 0.92 (0.89-0.95), respectively. Between 5 and 30 kPa, the log-HR increased as a monotonic function of LSM. The predictive value of LSM was stable in time. LSM improved the prognostic ability of biochemical response criteria, fibrosis scores, and prognostic scores. The 8 kPa and 15 kPa cut-offs optimally separated low-, medium-, and high-risk groups. Forty percent of patients were at medium to high risk according to LSM. CONCLUSIONS: LSM by VCTE is a major, independent, validated predictor of PBC outcome. Its value as a surrogate endpoint for clinical benefit in PBC should be considered. LAY SUMMARY: Primary biliary cholangitis (PBC) is a chronic autoimmune disease, wherein the body's immune system mistakenly attacks the bile ducts. PBC progresses gradually, so surrogate markers (markers that predict clinically relevant outcomes like the need for a transplant or death long before the event occurs) are often needed to expedite the drug development and approval process. Herein, we show that liver stiffness measurement is a strong predictor of clinical outcomes and could be a useful surrogate endpoint in PBC trials.

Clinical treatment of cholangiocarcinoma: an updated comprehensive review.

Cholangiocarcinoma (CCA) is a heterogeneous group of neoplasms of the bile ducts and represents the second most common hepatic cancer after hepatocellular carcinoma; it is sub-classified as intrahepatic cholangiocarcinoma (iCCA) and extrahepatic cholangiocarcinoma (eCCA), the latter comprising both perihilar cholangiocarcinoma (pCCA or Klatskin tumor), and distal cholangiocarcinoma (dCCA). The global incidence of CCA has increased worldwide in recent decades. Chronic inflammation of biliary epithelium and bile stasis represent the main risk factors shared by all CCA sub-types. When feasible, liver resection is the treatment of choice for CCA, followed by systemic chemotherapy with capecitabine. Liver transplants represent a treatment option in patients with very early iCCA, in referral centers only. CCA diagnosis is often performed at an advanced stage when CCA is unresectable. In this setting, systemic chemotherapy with gemcitabine and cisplatin represents the first treatment option, but the prognosis remains poor. In order to ameliorate patients' survival, new drugs have been studied in the last few years. Target therapies are directed against different molecules, which are altered in CCA cells. These therapies have been studied as second-line therapy, alone or in combination with chemotherapy. In the same setting, the immune checkpoints inhibitors targeting programmed death 1 (PD-1), programmed death-ligand 1 (PD-L1), cytotoxic T-lymphocyte antigen-4 (CTLA-4), have been proposed, as well as cancer vaccines and adoptive cell therapy (ACT). These experimental treatments showed promising results and have been proposed as second- or third-line treatment, alone or in combination with chemotherapy or target therapies.

The Role of Epigenetics in Primary Biliary Cholangitis.

Primary Biliary Cholangitis (PBC) is a rare autoimmune disease of the liver, affecting mostly females. There is evidence that epigenetic changes have a pathogenic role in PBC. Epigenetic modifications are related to methylation of CpG DNA islands, post-translational modifications of histone proteins, and non-coding RNAs. In PBC, there are data showing a dysregulation of all these levels, especially in immune cells. In addition, epigenetics seems to be involved in complex phenomena such as X monosomy or abnormalities in the process of X chromosome inactivation, which have been reported in PBC and appear to influence its sex imbalance and pathogenesis. We review here historical data on epigenetic modifications in PBC, present new data, and discuss possible links among X-chromosome abnormalities at a genetic and epigenetic level, PBC pathogenesis, and PBC sex imbalance.

Systematic review-pancreatic involvement in inflammatory bowel disease.

BACKGROUND: Inflammatory bowel disease (IBD) is a chronic inflammatory immune-mediated disorder of the gut with frequent extra-intestinal complications. Pancreatic involvement in IBD is not uncommon and comprises a heterogeneous group of conditions, including acute pancreatitis (AP), chronic pancreatitis (CP), autoimmune pancreatitis (AIP) and pancreatic exocrine insufficiency (PEI); however, data on such an association remain sparse and heterogeneous. METHOD: PubMed/MEDLINE and EMBASE databases were searched for studies investigating pancreatic involvement in patients with IBD. RESULTS: Four thousand one hundred and twenty-one records were identified and 547 screened; finally, 124 studies were included in the review. AP is the most frequent pancreatic manifestation in IBD; the majority of AP cases in IBD are due to gallstones and drugs but cases of idiopathic AP are increasingly reported. AIP is a rare disease, but a strong association with IBD has been demonstrated, especially for type 2 and ulcerative colitis. The pathogenetic link between IBD and AIP remains unclear, but an immune-mediated pathway seems plausible. An association between CP and PEI with IBD has also been suggested, but data are to date scarce and conflicting. CONCLUSION: This is the first systematic review of the association between IBD and pancreatic diseases. Gallstones and drugs should be considered the most probable causes of AP in IBD, with type 2 AIP also being possible.

X marks the spot in autoimmunity.

INTRODUCTION: Autoimmune diseases mostly affect females. Besides hormones, several factors related to chromosome X have been called in action to explain this sex predominance. AREAS COVERED: This paper provides an overview on the role of chromosome X (chrX) in explaining why females have higher susceptibility to autoimmunity. The work outlines some essential concepts regarding chrX inactivation, escape from chrX inactivation and the evolutionary history of chrX. In addition, we will discuss the concept of gene escape in immune cells, with examples related to specific X-linked genes and autoimmune diseases. EXPERT OPINION: There is growing evidence that many genes present on chrX escape inactivation, and some of them have significant immune-mediated functions. In immune cells of female individuals the escape of these genes is not constant, but the knowledge of the mechanisms controlling this plasticity are not completely understood. Future studies aimed at the characterization of these modifications at single-cell resolution, together with conformational 3D studies of the inactive X chromosome, will hopefully help to fill this gap of knowledge.

An update on novel pharmacological agents for primary sclerosing cholangitis.

INTRODUCTION: Primary sclerosing cholangitis (PSC) is a rare chronic cholestatic disease with heterogeneous phenotypes that may lead to liver transplantation and/or end-stage liver disease. Its multifactorial etiopathogenesis remains uncertain, but gut-liver axis and bile composition and excretion are widely demonstrated to influence the immune-mediated fibrogenic reactive cascade. AREAS COVERED: Different experimental therapeutic options are under investigation, mainly aiming at modulating bile acids excretion, limiting inflammatory-cascade reactions, and changing intestinal microbiota composition; none of them yet demonstrated to prolong transplant-free survival.This review provides a comprehensive description of the experimental drugs recently tested and/or currently under investigation. A bibliographical search was performed in PubMed, MEDLINE, EMBASE, OVID, and clinicaltrial.gov until July 2021. EXPERT OPINION: The heterogeneity and poor prevalence of PSC, its uncertain pathophysiology, and the lack of surrogate endpoints are the major challenges in drug discovery. Strategies that synergistically target microbiota, bile acids, and liver fibrosis are needed. In parallel, we must enhance biomarkers discovery to develop surrogate endpoints, as biochemical markers' fluctuations over the time hamper their effectiveness. Magnetic resonance cholangiopancreatography tools that accurately measure bile duct changes represent a potential marker for disease monitoring. A collaboration between academia, research consortia, patient's associations, and industry is required.

Effects of immunosuppressive drugs on COVID-19 severity in patients with autoimmune hepatitis.

BACKGROUND: We investigated associations between baseline use of immunosuppressive drugs and severity of Coronavirus Disease 2019 (COVID-19) in autoimmune hepatitis (AIH). PATIENTS AND METHODS: Data of AIH patients with laboratory confirmed COVID-19 were retrospectively collected from 15 countries. The outcomes of AIH patients who were on immunosuppression at the time of COVID-19 were compared to patients who were not on AIH medication. The clinical courses of COVID-19 were classified as (i)-no hospitalization, (ii)-hospitalization without oxygen supplementation, (iii)-hospitalization with oxygen supplementation by nasal cannula or mask, (iv)-intensive care unit (ICU) admission with non-invasive mechanical ventilation, (v)-ICU admission with invasive mechanical ventilation or (vi)-death and analysed using ordinal logistic regression. RESULTS: We included 254 AIH patients (79.5%, female) with a median age of 50 (range, 17-85) years. At the onset of COVID-19, 234 patients (92.1%) were on treatment with glucocorticoids (n = 156), thiopurines (n = 151), mycophenolate mofetil (n = 22) or tacrolimus (n = 16), alone or in combinations. Overall, 94 (37%) patients were hospitalized and 18 (7.1%) patients died. Use of systemic glucocorticoids (adjusted odds ratio [aOR] 4.73, 95% CI 1.12-25.89) and thiopurines (aOR 4.78, 95% CI 1.33-23.50) for AIH was associated with worse COVID-19 severity, after adjusting for age-sex, comorbidities and presence of cirrhosis. Baseline treatment with mycophenolate mofetil (aOR 3.56, 95% CI 0.76-20.56) and tacrolimus (aOR 4.09, 95% CI 0.69-27.00) were also associated with more severe COVID-19 courses in a smaller subset of treated patients. CONCLUSION: Baseline treatment with systemic glucocorticoids or thiopurines prior to the onset of COVID-19 was significantly associated with COVID-19 severity in patients with AIH.

Machine learning in primary biliary cholangitis: A novel approach for risk stratification.

BACKGROUND & AIMS: Machine learning (ML) provides new approaches for prognostication through the identification of novel subgroups of patients. We explored whether ML could support disease sub-phenotyping and risk stratification in primary biliary cholangitis (PBC). METHODS: ML was applied to an international dataset of PBC patients. The dataset was split into a derivation cohort (training set) and a validation cohort (validation set), and key clinical features were analysed. The outcome was a composite of liver-related death or liver transplantation. ML and standard survival analysis were performed. RESULTS: The training set was composed of 11,819 subjects, while the validation set was composed of 1,069 subjects. ML identified four clusters of patients characterized by different phenotypes and long-term prognosis. Cluster 1 (n = 3566) included patients with excellent prognosis, whereas Cluster 2 (n = 3966) consisted of individuals at worse prognosis differing from Cluster 1 only for albumin levels around the limit of normal. Cluster 3 (n = 2379) included young patients with florid cholestasis and Cluster 4 (n = 1908) comprised advanced cases. Further sub-analyses on the dynamics of albumin within the normal range revealed that ursodeoxycholic acid-induced increase of albumin >1.2 x lower limit of normal (LLN) is associated with improved transplant-free survival. CONCLUSIONS: Unsupervised ML identified four novel groups of PBC patients with different phenotypes and prognosis and highlighted subtle variations of albumin within the normal range. Therapy-induced increase of albumin >1.2 x LLN should be considered a treatment goal.