Machine Learning-Based Models for Advanced Fibrosis and Cirrhosis Diagnosis in Chronic Hepatitis B Patients With Hepatic Steatosis.

BACKGROUND AND AIMS: The global rise of chronic hepatitis B (CHB) superimposed on hepatic steatosis (HS) warrants noninvasive, precise tools for assessing fibrosis progression. This study leveraged machine learning (ML) to develop diagnostic models for advanced fibrosis and cirrhosis in this patient population. METHODS: Treatment-naive CHB patients with concurrent HS who underwent liver biopsy in 10 medical centers were enrolled as a training cohort and an independent external validation cohort (NCT05766449). Six ML models were implemented to predict advanced fibrosis and cirrhosis. The final models, derived from SHAP (Shapley Additive exPlanations), were compared with Fibrosis-4 Index, nonalcoholic fatty liver disease Fibrosis Score, and aspartate aminotransferase-to-platelet ratio index using the area under receiver-operating characteristic curve (AUROC) and decision curve analysis (DCA). RESULTS: Of 1,198 eligible patients, the random forest model achieved AUROCs of 0.778 (95% confidence interval [CI], 0.749-0.807) for diagnosing advanced fibrosis (random forest advanced fibrosis model) and 0.777 (95% CI, 0.748-0.806) for diagnosing cirrhosis (random forest cirrhosis model) in the training cohort, and maintained high AUROCs in the validation cohort. In the training cohort, the random forest advanced fibrosis model obtained an AUROC of 0.825 (95% CI, 0.787-0.862) in patients with hepatitis B virus DNA ≥105 IU/mL, and the random forest cirrhosis model had an AUROC of 0.828 (95% CI, 0.774-0.883) in female patients. The 2 models outperformed Fibrosis-4 Index, nonalcoholic fatty liver disease Fibrosis Score, and aspartate aminotransferase-to-platelet ratio index in the training cohort, and also performed well in the validation cohort. CONCLUSIONS: The random forest models provide reliable, noninvasive tools for identifying advanced fibrosis and cirrhosis in CHB patients with concurrent HS, offering a significant advancement in the comanagement of the 2 diseases. CLINICALTRIALS: gov, Number: NCT05766449.

Liver inflammation activity in patients with autoimmune hepatitis with normal alanine aminotransferase and immunoglobulin G levels.

Background and aims: Normal serum transaminases and immunoglobulin G (IgG) levels are surrogate markers for hepatic histologic disease activity in autoimmune hepatitis (AIH). This study aimed to evaluate liver inflammation in patients with AIH with normal serum alanine aminotransferase (ALT) and IgG levels. Methods: Two hundred and five AIH patients who underwent liver biopsy in four medical centers were included. Logistic regression analysis was used to identify risk factors associated with advanced inflammation. Results: One hundred and thirty-one (63.9 %) AIH patients had advanced liver inflammation, and 108 (52.7 %) patients had advanced liver fibrosis. 60.0 % of patients with normal ALT and 51.7 % of patients with normal ALT and IgG had advanced inflammation. However, 76.7 % and 35.0 % of patients with or without advanced fibrosis with normal ALT had advanced inflammation, while the corresponding proportions of advanced inflammation were 78.6 % and 26.7 % in patients with normal ALT and IgG, respectively. Moreover, 81.0 % and 44.8 % of patients with and without cirrhosis with normal ALT had advanced inflammation, while the corresponding proportions were 83.3 % and 29.4 % in patients with normal ALT and IgG, respectively. Red cell distribution width (OR = 1.325, 95%CI 1.045-1.681, P = 0.020) and PT (OR = 1.514, 95%CI 1.138-2.014, P = 0.004) were independent factors associated with advanced inflammation. Conclusions: High proportion of advanced inflammation was found in AIH patients with normal ALT and IgG levels despite without advanced fibrosis. Although using non-invasive methods may contribute to rule out liver fibrosis in AIH patients with normal ALT and IgG levels, liver biopsy is encouraged to assess liver inflammation.

Deciphering lung granulomas in HIV & TB co-infection: unveiling macrophages aggregation with IL6R/STAT3 activation.

Tuberculosis (TB) remains a leading cause of mortality among individuals coinfected with HIV, characterized by progressive pulmonary inflammation. Despite TB's hallmark being focal granulomatous lung lesions, our understanding of the histopathological features and regulation of inflammation in HIV & TB coinfection remains incomplete. In this study, we aimed to elucidate these histopathological features through an immunohistochemistry analysis of HIV & TB co-infected and TB patients, revealing marked differences. Notably, HIV & TB granulomas exhibited aggregation of CD68 + macrophage (Mφ), while TB lesions predominantly featured aggregation of CD20+ B cells, highlighting distinct immune responses in coinfection. Spatial transcriptome profiling further elucidated CD68+ Mφ aggregation in HIV & TB, accompanied by activation of IL6 pathway, potentially exacerbating inflammation. Through multiplex immunostaining, we validated two granuloma types in HIV & TB versus three in TB, distinguished by cell architecture. Remarkably, in the two types of HIV & TB granulomas, CD68 + Mφ highly co-expressed IL6R/pSTAT3, contrasting TB granulomas' high IFNGRA/SOCS3 expression, indicating different signaling pathways at play. Thus, activation of IL6 pathway may intensify inflammation in HIV & TB-lungs, while SOCS3-enriched immune microenvironment suppresses IL6-induced over-inflammation in TB. These findings provide crucial insights into HIV & TB granuloma formation, shedding light on potential therapeutic targets, particularly for granulomatous pulmonary under HIV & TB co-infection. Our study emphasizes the importance of a comprehensive understanding of the immunopathogenesis of HIV & TB coinfection and suggests potential avenues for targeting IL6 signaling with SOCS3 activators or anti-IL6R agents to mitigate lung inflammation in HIV & TB coinfected individuals.

Cell-type-specific expression analysis of liver transcriptomics with clinical parameters to decipher the cause of intrahepatic inflammation in chronic hepatitis B.

Functional cure for chronic hepatitis B (CHB) remains challenging due to the lack of direct intervention methods for hepatic inflammation. Multi-omics research offers a promising approach to understand hepatic inflammation mechanisms in CHB. A Bayesian linear model linked gene expression with clinical parameters, and population-specific expression analysis (PSEA) refined bulk gene expression into specific cell types across different clinical phases. These models were integrated into our analysis of key factors like inflammatory cells, immune activation, T cell exhaustion, chemokines, receptors, and interferon-stimulated genes (ISGs). Validation through multi-immune staining in liver specimens from CHB patients bolstered our findings. In CHB patients, increased gene expression related to immune cell activation and migration was noted. Marker genes of macrophages, T cells, immune-negative regulators, chemokines, and ISGs showed a positive correlation with serum alanine aminotransferase (ALT) levels but not hepatitis B virus DNA levels. The PSEA model confirmed T cells as the source of exhausted regulators, while macrophages primarily contributed to chemokine expression. Upregulated ISGs (ISG20, IFI16, TAP2, GBP1, PSMB9) in the hepatitis phase were associated with T cell and macrophage infiltration and positively correlated with ALT levels. Conversely, another set of ISGs (IFI44, ISG15, IFI44L, IFI6, MX1) mainly expressed by hepatocytes and B cells showed no correlation with ALT levels. Our study presents a multi-omics analysis integrating bulk transcriptomic, single-cell sequencing data, and clinical data from CHB patients to decipher the cause of intrahepatic inflammation in CHB. The findings confirm that macrophages secrete chemokines like CCL20, recruiting exhausted T cells into liver tissue; concurrently, hepatocyte innate immunity is suppressed, hindering the antiviral effects of ISGs.

Histological features of chronic hepatitis B patients with normal alanine aminotransferase according to different criteria.

BACKGROUND: The upper limits of normal (ULNs) for alanine aminotransferase (ALT) are different among international guidelines for chronic hepatitis B (CHB). We aimed to investigate the proportion of significant histological disease in Asian patients with CHB with detectable hepatitis B virus (HBV) DNA under diverse ALT ULNs. METHODS: Consecutive patients with CHB and detectable HBV DNA who underwent liver biopsy were retrospectively included from four tertiary hospitals. Above grade 2 inflammation and stage 2 fibrosis were defined as significant inflammation and significant fibrosis, respectively. Significant histological disease was defined as above grade 2 inflammation or stage 2 fibrosis. RESULTS: Among the 414 patients with detectable HBV DNA and normal ALT, the proportion of those with significant histological disease was lower (59.7%) according to the ULN for ALT at 30/19 U/L (male/female), while the corresponding proportions were 66.7% and 62.3% according to the ULNs of 40 U/L and 35/25 U/L (male/female), respectively. In patients with detectable HBV DNA and normal ALT levels without significant fibrosis, the proportions of significant inflammation were comparable among different ULNs of ALT at 40 U/L (30.7%), 35/25 U/L (27.3%) and 30/19 U/L (25.0%). The proportion of significant histological disease was significantly lower in patients with normal ALT for 2 determinations at least 6 months apart compared to patients with normal ALT once. CONCLUSIONS: Although a more stringent ALT ULN may reduce the risk of the presence of significant histological disease in patients with detectable HBV DNA, the rates of significant histological disease remain high. Persistently normal ALT levels are more important for excluding patients with CHB with a high probability of significant histological disease.

Development of a machine learning-based model to predict hepatic inflammation in chronic hepatitis B patients with concurrent hepatic steatosis: a cohort study.

Background: With increasingly prevalent coexistence of chronic hepatitis B (CHB) and hepatic steatosis (HS), simple, non-invasive diagnostic methods to accurately assess the severity of hepatic inflammation are needed. We aimed to build a machine learning (ML) based model to detect hepatic inflammation in patients with CHB and concurrent HS. Methods: We conducted a multicenter, retrospective cohort study in China. Treatment-naive CHB patients with biopsy-proven HS between April 2004 and September 2022 were included. The optimal features for model development were selected by SHapley Additive explanations, and an ML algorithm with the best accuracy to diagnose moderate to severe hepatic inflammation (Scheuer's system ≥ G3) was determined and assessed by decision curve analysis (DCA) and calibration curve. This study is registered with ClinicalTrials.gov (NCT05766449). Findings: From a pool of 1,787 treatment-naive patients with CHB and HS across eleven hospitals, 689 patients from nine of these hospitals were chosen for the development of the diagnostic model. The remaining two hospitals contributed to two independent external validation cohorts, comprising 509 patients in validation cohort 1 and 589 in validation cohort 2. Eleven features regarding inflammation, hepatic and metabolic functions were identified. The gradient boosting classifier (GBC) model showed the best performance in predicting moderate to severe hepatic inflammation, with an area under the receiver operating characteristic curve (AUROC) of 0.86 (95% CI 0.83-0.88) in the training cohort, and 0.89 (95% CI 0.86-0.92), 0.76 (95% CI 0.73-0.80) in the first and second external validation cohorts, respectively. A publicly accessible web tool was generated for the model. Interpretation: Using simple parameters, the GBC model predicted hepatic inflammation in CHB patients with concurrent HS. It holds promise for guiding clinical management and improving patient outcomes. Funding: This research was supported by the National Natural Science Foundation of China (No. 82170609, 81970545), Natural Science Foundation of Shandong Province (Major Project) (No. ZR2020KH006), Natural Science Foundation of Jiangsu Province (No.BK20231118), Tianjin Key Medical Discipline (Specialty), Construction Project, TJYXZDXK-059B, Tianjin Health Science and Technology Project key discipline special, TJWJ2022XK034, and Research project of Chinese traditional medicine and Chinese traditional medicine combined with Western medicine of Tianjin municipal health and Family Planning Commission (2021022).