Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information.

AIMS: Metabolic dysfunction Associated Steatotic Liver Disease (MASLD) outcomes such as MASH (metabolic dysfunction associated steatohepatitis), fibrosis and cirrhosis are ordinarily determined by resource-intensive and invasive biopsies. We aim to show that routine clinical tests offer sufficient information to predict these endpoints. METHODS: Using the LITMUS Metacohort derived from the European NAFLD Registry, the largest MASLD dataset in Europe, we create three combinations of features which vary in degree of procurement including a 19-variable feature set that are attained through a routine clinical appointment or blood test. This data was used to train predictive models using supervised machine learning (ML) algorithm XGBoost, alongside missing imputation technique MICE and class balancing algorithm SMOTE. Shapley Additive exPlanations (SHAP) were added to determine relative importance for each clinical variable. RESULTS: Analysing nine biopsy-derived MASLD outcomes of cohort size ranging between 5385 and 6673 subjects, we were able to predict individuals at training set AUCs ranging from 0.719-0.994, including classifying individuals who are At-Risk MASH at an AUC = 0.899. Using two further feature combinations of 26-variables and 35-variables, which included composite scores known to be good indicators for MASLD endpoints and advanced specialist tests, we found predictive performance did not sufficiently improve. We are also able to present local and global explanations for each ML model, offering clinicians interpretability without the expense of worsening predictive performance. CONCLUSIONS: This study developed a series of ML models of accuracy ranging from 71.9-99.4% using only easily extractable and readily available information in predicting MASLD outcomes which are usually determined through highly invasive means.

Thyroid function, adipokines and mitokines in metabolic dysfunction-associated steatohepatitis: A multi-centre biopsy-based observational study.

BACKGROUND AND AIMS: Thyroid axis is currently under investigation as a therapeutic target in metabolic dysfunction-associated steatotic liver disease (MASLD). Thyroid function was examined herein in the full spectrum of disease. METHODS: Subjects were recruited and had liver biopsies in two Gastroenterology-Hepatology Clinics (Greece and Australia) and one Bariatric-Metabolic Surgery Clinic (Italy). The main working sample was n = 677 subjects with MASLD after excluding subjects with abnormal free thyroxine levels. Participants were classified according to thyroid-stimulating hormone (TSH) standard criteria: Subclinical hyperthyroidism (<0.4 uIU/mL); Euthyroidism with relatively low (0.4 to <2.5 uIU/mL); euthyroidism with relatively high (2.5-4.0 uIU/mL); subclinical hypothyroidism (>4 uIU/mL). RESULTS: TSH as a continuous variable was positively associated with significant fibrosis (F ≥ 2), metabolic dysfunction-associated steatohepatitis (MASH) and at-risk MASH. Subclinical hypothyroidism was associated with fibrosis F ≥ 2 (odds ratio [OR] = 3.47, 95% confident interval [CI] [1.50, 8.05], p = .02), MASH (OR = 3.44, 95% CI [1.48, 7.98] p = .001) and at-risk MASH (OR = 3.88, 95% CI [1.76, 8.55], p = .001), before and after controlling for adiposity, central obesity, and insulin resistance. When leptin, adiponectin, or growth differentiation factor-15 were examined as moderators, significance was lost. Sex-specific analysis revealed a strong association between TSH and the presence of significant fibrosis among women, eliminated only when adipokines/mitokines were adjusted for. Restricted cubic spline analysis revealed associations between TSH and liver outcomes (p-values < .01) with inflection points for fibrosis F ≥ 2 being 2.49, for MASH being 2.67 and for at-risk MASH being 6.96. CONCLUSIONS: These observations provide support for studies on the administration of thyroid hormone in MASLD therapeutics for subclinical hypothyroidism and liver-specific thyroid receptor agonists for subjects across the TSH continuum.

Circulating hormones in biopsy-proven steatotic liver disease and steatohepatitis: A Multicenter Observational Study.

BACKGROUND: The role of metabolic/inflammatory hormonal systems in metabolic dysfunction associated steatotic liver disease (MASLD) remains to be fully elucidated. PURPOSE: To report the levels of the novel total and H-specific growth differentiation factor-15 (GDF-15) and other established hormonal systems and to describe hormonal patterns in controls and patients with MASLD and its stages. METHODS: This is a multicenter study from two Gastroenterology-Hepatology Departments (Greece and Australia) and one Bariatric-Metabolic Surgery Department (Italy). Overall, n = 455 serum samples of patients with biopsy-proven MASLD (n = 374) and Controls (n = 81) were recruited. RESULTS: We report for the first time that total and H-specific GDF-15 levels are higher in MASLD, at-risk metabolic dysfunction associated steatohepatitis (MASH), and severe fibrosis than in Controls. In addition, follistatin-like-3 (FSTL-3), free insulin-like growth factor-1 (IGF-1), leptin, and insulin levels were higher in MASLD patients than in Controls, while adiponectin levels were lower in MASLD subjects than in Controls. Activin-A, follistatin (FST), FSTL-3, and insulin levels significantly increased in severe fibrosis compared to no/mild fibrosis, while free IGF-1 decreased. In addition, adiponectin levels were lower in subjects without fibrosis vs. any fibrosis. Moreover, GDF-15 presented a strong positive association for the likelihood of having MASLD and at-risk MASH, while in adjusted analyses, FST and adiponectin showed inverse associations. Two different patterns of at-risk MASH were revealed through unsupervised analysis (total variation explained=54%). The most frequent pattern met in our sample (34.3%) was characterized by higher levels of total and H-specific GDF-15, follistatins, and activins, as well as low adiponectin levels. The second pattern revealed was characterized by high levels of free IGF-1, insulin, and leptin, with low levels of activin-A and adiponectin. Similar patterns were also generated in the case of overall MASLD. CONCLUSIONS: Total and H-specific GDF-15 levels increase as MASLD severity progresses. FSTL-3, free IGF-1, leptin, and insulin are also higher, whereas adiponectin and activin-A levels are lower in the MASLD group than in Controls. Hormonal systems, including GDF-15, may not only be involved in the pathophysiology but could also prove useful for the diagnostic workup of MASLD and its stages and may potentially be of therapeutic value.

The first external validation of the Dallas steatosis index in biopsy-proven Non-alcoholic fatty liver Disease: A multicenter study.

AIMS: A new non-invasive tool (NIT) for non-alcoholic fatty liver disease (NAFLD) proposed in 2022 by the multi-ethnic Dallas Heart Study, i.e. the Dallas Steatosis Index (DSI), was validated herein using for the first time the gold standard i.e. liver biopsy-proven NAFLD. METHODS: This is a multicenter study based on samples and data from two Gastroenterology-Hepatology Clinics (Greece and Australia) and one Bariatric-Metabolic Surgery Clinic (Italy). Overall, n = 455 patients with biopsy-proven NAFLD (n = 374) and biopsy-proven controls (n = 81) were recruited. RESULTS: The ability of DSI to correctly classify participants as NAFLD or controls was very good, reaching an Area Under the Curve (AUC) = 0.887. The cut-off point that could best differentiate the presence vs. absence of NAFLD corresponded to DSI = 0.0 (risk threshold: 50% | Sensitivity: 0.88; Positive Predictive Value (PPV): 93.0%; F1-score = 0.91). DSI demonstrated significantly better performance characteristics than other liver steatosis indexes. Decision curve analysis revealed that the benefit of DSI as a marker to indicate the need for invasive liver assessment was confirmed only when higher DSI values, i.e. ≥ 1.4, were used as risk thresholds. DSI performance to differentiate disease progression was inadequate (all AUCs < 0.700). CONCLUSIONS: DSI is more useful for disease screening (NAFLD vs. controls) than to differentiate diseases stages or progression. The value of any inclusion of DSI to guidelines needs to be further studied.

Liver biopsy-based validation, confirmation and comparison of the diagnostic performance of established and novel non-invasive steatotic liver disease indexes: Results from a large multi-center study.

BACKGROUND: Non-invasive tools (NIT) for metabolic-dysfunction associated liver disease (MASLD) screening or diagnosis need to be thoroughly validated using liver biopsies. PURPOSE: To externally validate NITs designed to differentiate the presence or absence of liver steatosis as well as more advanced disease stages, to confirm fully validated indexes (n = 7 NITs), to fully validate partially validated indexes (n = 5 NITs), and to validate for the first time one new index (n = 1 NIT). METHODS: This is a multi-center study from two Gastroenterology-Hepatology Departments (Greece and Australia) and one Bariatric-Metabolic Surgery Department (Italy). Overall, n = 455 serum samples of patients with biopsy-proven MASLD (n = 374, including 237 patients with metabolic-dysfunction associated steatohepatitis (MASH)) and Controls (n = 81) were recruited. A complete validation analysis was performed to differentiate the presence of MASLD vs. Controls, MASH vs. metabolic-dysfunction associated steatotic liver (MASL), histological features of MASH, and fibrosis stages. RESULTS: The index of NASH (ION) demonstrated the highest differentiation ability for the presence of MASLD vs. Controls, with the area under the curve (AUC) being 0.894. For specific histological characterization of MASH, no NIT demonstrated adequate performance, while in the case of specific features of MASH, such as hepatocellular ballooning and lobular inflammation, ION demonstrated the best performance with AUC being close to or above 0.850. For fibrosis (F) classification, the highest AUC was reached by the aspartate aminotransferase to platelet ratio index (APRI) being ~0.850 yet only with the potential to differentiate the severe fibrosis stages (F3, F4) vs. mild or moderate fibrosis (F0-2) with an AUC > 0.900 in patients without T2DM. When we excluded patients with morbid obesity, the differentiation ability of APRI was improved, reaching AUC = 0.802 for differentiating the presence of fibrosis F2-4 vs. F0-1. The recommended by current guidelines index FIB-4 seemed to differentiate adequately between severe (i.e., F3-4) and mild or moderate fibrosis (F0-2) with an AUC = 0.820, yet this was not the case when FIB-4 was used to classify patients with fibrosis F2-4 vs. F0-1. Trying to improve the predictive value of all NITs, using Youden's methodology, to optimize the suggested cut-off points did not materially improve the results. CONCLUSIONS: The validation of currently available NITs using biopsy-proven samples provides new evidence for their ability to differentiate between specific disease stages, histological features, and, most importantly, fibrosis grading. The overall performance of the examined NITs needs to be further improved for applications in the clinic.

Machine learning algorithm improves the detection of NASH (NAS-based) and at-risk NASH: A development and validation study.

BACKGROUND AND AIMS: Detecting NASH remains challenging, while at-risk NASH (steatohepatitis and F≥ 2) tends to progress and is of interest for drug development and clinical application. We developed prediction models by supervised machine learning techniques, with clinical data and biomarkers to stage and grade patients with NAFLD. APPROACH AND RESULTS: Learning data were collected in the Liver Investigation: Testing Marker Utility in Steatohepatitis metacohort (966 biopsy-proven NAFLD adults), staged and graded according to NASH CRN. Conditions of interest were the clinical trial definition of NASH (NAS ≥ 4;53%), at-risk NASH (NASH with F ≥ 2;35%), significant (F ≥ 2;47%), and advanced fibrosis (F ≥ 3;28%). Thirty-five predictors were included. Missing data were handled by multiple imputations. Data were randomly split into training/validation (75/25) sets. A gradient boosting machine was applied to develop 2 models for each condition: clinical versus extended (clinical and biomarkers). Two variants of the NASH and at-risk NASH models were constructed: direct and composite models.Clinical gradient boosting machine models for steatosis/inflammation/ballooning had AUCs of 0.94/0.79/0.72. There were no improvements when biomarkers were included. The direct NASH model produced AUCs (clinical/extended) of 0.61/0.65. The composite NASH model performed significantly better (0.71) for both variants. The composite at-risk NASH model had an AUC of 0.83 (clinical and extended), an improvement over the direct model. Significant fibrosis models had AUCs (clinical/extended) of 0.76/0.78. The extended advanced fibrosis model (0.86) performed significantly better than the clinical version (0.82). CONCLUSIONS: Detection of NASH and at-risk NASH can be improved by constructing independent machine learning models for each component, using only clinical predictors. Adding biomarkers only improved the accuracy of fibrosis.

Biomarkers for staging fibrosis and non-alcoholic steatohepatitis in non-alcoholic fatty liver disease (the LITMUS project): a comparative diagnostic accuracy study.

BACKGROUND: The reference standard for detecting non-alcoholic steatohepatitis (NASH) and staging fibrosis-liver biopsy-is invasive and resource intensive. Non-invasive biomarkers are urgently needed, but few studies have compared these biomarkers in a single cohort. As part of the Liver Investigation: Testing Marker Utility in Steatohepatitis (LITMUS) project, we aimed to evaluate the diagnostic accuracy of 17 biomarkers and multimarker scores in detecting NASH and clinically significant fibrosis in patients with non-alcoholic fatty liver disease (NAFLD) and identify their optimal cutoffs as screening tests in clinical trial recruitment. METHODS: This was a comparative diagnostic accuracy study in people with biopsy-confirmed NAFLD from 13 countries across Europe, recruited between Jan 6, 2010, and Dec 29, 2017, from the LITMUS metacohort of the prospective European NAFLD Registry. Adults (aged ≥18 years) with paired liver biopsy and serum samples were eligible; those with excessive alcohol consumption or evidence of other chronic liver diseases were excluded. The diagnostic accuracy of the biomarkers was expressed as the area under the receiver operating characteristic curve (AUC) with liver histology as the reference standard and compared with the Fibrosis-4 index for liver fibrosis (FIB-4) in the same subgroup. Target conditions were the presence of NASH with clinically significant fibrosis (ie, at-risk NASH; NAFLD Activity Score ≥4 and F≥2) or the presence of advanced fibrosis (F≥3), analysed in all participants with complete data. We identified thres holds for each biomarker for reducing the number of biopsy-based screen failures when recruiting people with both NASH and clinically significant fibrosis for future trials. FINDINGS: Of 1430 participants with NAFLD in the LITMUS metacohort with serum samples, 966 (403 women and 563 men) were included after all exclusion criteria had been applied. 335 (35%) of 966 participants had biopsy-confirmed NASH and clinically significant fibrosis and 271 (28%) had advanced fibrosis. For people with NASH and clinically significant fibrosis, no single biomarker or multimarker score significantly reached the predefined AUC 0·80 acceptability threshold (AUCs ranging from 0·61 [95% CI 0·54-0·67] for FibroScan controlled attenuation parameter to 0·81 [0·75-0·86] for SomaSignal), with accuracy mostly similar to FIB-4. Regarding detection of advanced fibrosis, SomaSignal (AUC 0·90 [95% CI 0·86-0·94]), ADAPT (0·85 [0·81-0·89]), and FibroScan liver stiffness measurement (0·83 [0·80-0·86]) reached acceptable accuracy. With 11 of 17 markers, histological screen failure rates could be reduced to 33% in trials if only people who were marker positive had a biopsy for evaluating eligibility. The best screening performance for NASH and clinically significant fibrosis was observed for SomaSignal (number needed to test [NNT] to find one true positive was four [95% CI 4-5]), then ADAPT (six [5-7]), MACK-3 (seven [6-8]), and PRO-C3 (nine [7-11]). INTERPRETATION: None of the single markers or multimarker scores achieved the predefined acceptable AUC for replacing biopsy in detecting people with both NASH and clinically significant fibrosis. However, several biomarkers could be applied in a prescreening strategy in clinical trial recruitment. The performance of promising markers will be further evaluated in the ongoing prospective LITMUS study cohort. FUNDING: The Innovative Medicines Initiative 2 Joint Undertaking.

Clinical and patient-reported outcome profile of patients with hepatitis B viral infection from the Global Liver Registry™.

Chronic hepatitis B (CHB) infection is one of the most common causes of cirrhosis and liver cancer worldwide. Our aim was to assess clinical and patient-reported outcome (PRO) profile of CHB patients from different regions of the world using the Global Liver Registry. The CHB patients seen in real-world practices are being enrolled in the Global Liver Registry. Clinical and PRO (FACIT-F, CLDQ, WPAI) data were collected and compared to baseline data from CHB controls from clinical trials. The study included 1818 HBV subjects (48 ± 13 years, 58% male, 14% advanced fibrosis, 7% cirrhosis) from 15 countries in 6/7 Global Burden of Disease super-regions. The rates of advanced fibrosis varied (3-24%). The lowest PRO scores across multiple domains were in HBV subjects from the Middle East/North Africa (MENA), the highest - Southeast/East and South Asia. Subjects with advanced fibrosis had PRO impairment in 3 CLDQ domains, Activity of WPAI (p < 0.05). HBV subjects with superimposed fatty liver had more PRO impairments. In multivariate analysis adjusted for location, predictors of PRO impairment in CHB included female sex, advanced fibrosis, and non-hepatic comorbidities (p < 0.05). In comparison to Global Liver Registry patients, 242 controls from clinical trials had better PRO scores (Abdominal, Emotional, and Systemic scores of CLDQ, all domains of WPAI) (p < 0.05). In multivariate analysis with adjustment for location and clinicodemographic parameters, the associations of PROs with the enrollment setting (real-life Global Liver Registry vs. clinical trials) were no longer significant (all p > 0.10). The clinico-demographic portrait of CHB patients varies across regions of the world and enrollment settings. Advanced fibrosis and non-hepatic comorbidities are independently associated with PRO impairment in CHB patients.

Acute and Late Rectal Toxicity Following Hypofractionated Radiotherapy in Patients With Prostate Cancer: Results of a Prospective Study.

BACKGROUND/AIM: Previous randomized clinical trials have shown that moderate hypofractionation has a non-inferior or even superior efficacy to conventionally fractionated external beam radiation therapy (EBRT) in low and intermediate-risk prostate cancer. We herein aimed to evaluate the acute and late gastrointestinal (GI) toxicity of hypofractionated radiotherapy (HRT) in a real-world setting. PATIENTS AND METHODS: Patients with intermediate-risk prostate adenocarcinoma eligible to receive HRT were prospectively enrolled. All patients were submitted to rectoscopy after completion of HRT, every three months after radiotherapy for the first year and every six months for the second year. Toxicity events were classified as acute, when presenting during radiotherapy or within the first three months following its completion, and as late when appearing three months to three years post-HRT. RESULTS: Twenty prostate cancer patients participated in this study and received 22 sessions of HRT (5 sessions a week; 2.75 Gy per session) and an overall dose of 60.5 Gy. None of our patients developed acute GI toxicity; late GI toxicity (RTOG/EORTC grade 3 rectal bleeding) was observed in 1 patient only (1/20, 5%), at 6- and 12-months post-HRT. No rectal mucosa damage was observed on follow-up rectoscopy in the acute phase in any of our patients; five patients (5/20, 25%) developed late telangiectasias. Vienna retroscopy score (VRS) was 1 in 4/5 patients (80%) and 2 in 1/5 (20%). CONCLUSION: Minimal radiation-induced rectal mucosal damage was observed in our patient population, and only as a late event, further attesting to the safety of HRT in this setting.

Final analysis of the international observational S-Collate study of peginterferon alfa-2a in patients with chronic hepatitis B.

BACKGROUND AND AIMS: Sustained off-treatment immune control is achievable in a proportion of patients with chronic hepatitis B treated with peginterferon alfa-2a. We evaluated on-treatment predictors of hepatitis B surface antigen (HBsAg) clearance 3 years after peginterferon alfa-2a treatment and determined the incidence of hepatocellular carcinoma. METHODS: A prospective, international, multicenter, observational study in patients with chronic hepatitis B who have been prescribed peginterferon alfa-2a (40KD) in a real-world setting. The primary endpoint was HBsAg clearance after 3 years' follow-up. RESULTS: The modified intention-to-treat population comprised 844 hepatitis B e antigen (HBeAg)-positive patients (540 [64%] completed 3 years' follow-up), and 872 HBeAg-negative patients (614 [70%] completed 3 years' follow-up). At 3 years' follow-up, HBsAg clearance rates in HBeAg-positive and HBeAg-negative populations, respectively, were 2% (16/844) and 5% (41/872) in the modified intention-to-treat population and 5% [16/328] and 10% [41/394] in those with available data. In HBeAg-positive patients with data, Week 12 HBsAg levels <1500, 1500-20,000, and >20,000 IU/mL were associated with HBsAg clearance rates at 3 years' follow-up of 11%, 1%, and 5%, respectively (Week 24 predictability was similar). In HBeAg-negative patients with available data, a ≥10% decline vs a <10% decline in HBsAg at Week 12 was associated with HBsAg clearance rates of 16% vs 4%. Hepatocellular carcinoma incidence was lower than REACH-B (Risk Estimation for Hepatocellular Carcinoma in Chronic Hepatitis B) model predictions. CONCLUSIONS: Sustained off-treatment immune control is achieved with peginterferon alfa-2a in a real-world setting. HBsAg clearance 3 years after completion of peginterferon alfa-2a can be predicted on the basis of on-treatment HBsAg kinetics.

The comprehensive outcomes of hepatitis C virus infection: A multi-faceted chronic disease.

Treatment of hepatitis C virus (HCV) infection has been revolutionized with the introduction of pangenotypic, interferon- and ribavirin-free regimens associated with high cure rates and a low side effect profile. Additionally, there is evidence that HCV cure reduces HCV complications, improves patient-reported outcomes and is cost-saving in most western countries in the long term. This is a review of the comprehensive burden of HCV and the value of eliminating HCV infection. With the introduction of the interferon-free all-oral, once a day pill treatment regimen for the cure of HCV, the potential to eliminate HCV by 2030 has become a possibility for some regions of the world. Nevertheless, there are barriers to screening, linkage to care, and treatment in many countries that must be overcome in order to reach this goal. In conclusion, globally, work must continue to ensure national policies are in place to support screening, linkage to care and affordable treatment in order to eliminate HCV.