Machine learning models are superior to noninvasive tests in identifying clinically significant stages of NAFLD and NAFLD-related cirrhosis

Diagnosis and non-invasive assessment of MASLD in type 2 diabetes and obesity

BACKGROUND

Metabolic dysfunction-associated steatotic liver disease (MASLD) is currently the most common chronic liver disease and an important cause of cirrhosis and hepatocellular carcinoma. It is strongly associated with type 2 diabetes and obesity. Because of the huge number of patients at risk of MASLD, it is imperative to use non-invasive tests appropriately.

AIMS

To provide a narrative review on the performance and limitations of non-invasive tests, with a special emphasis on the impact of diabetes and obesity.

METHODS

We searched PubMed and Cochrane databases for articles published from 1990 to August 2023.

RESULTS

Abdominal ultrasonography remains the primary method to diagnose hepatic steatosis, while magnetic resonance imaging proton density fat fraction is currently the gold standard to quantify steatosis. Simple fibrosis scores such as the Fibrosis-4 index are well suited as initial assessment in primary care and non-hepatology settings to rule out advanced fibrosis and future risk of liver-related complications. However, because of its low positive predictive value, an abnormal test should be followed by specific blood (e.g. Enhanced Liver Fibrosis score) or imaging biomarkers (e.g. vibration-controlled transient elastography and magnetic resonance elastography) of fibrosis. Some non-invasive tests of fibrosis appear to be less accurate in patients with diabetes. Obesity also affects the performance of abdominal ultrasonography and transient elastography, whereas magnetic resonance imaging may not be feasible in some patients with severe obesity.

CONCLUSIONS

This article highlights issues surrounding the clinical application of non-invasive tests for MASLD in patients with type 2 diabetes and obesity.

FibroScan-aspartate transaminase: A superior non-invasive model for diagnosing high-risk metabolic dysfunction-associated steatohepatitis

BACKGROUND

Non-alcoholic fatty liver disease (NAFLD) with hepatic histological NAFLD activity score ≥ 4 and fibrosis stage F ≥ 2 is regarded as “at risk” non-alcoholic steatohepatitis (NASH). Based on an international consensus, NAFLD and NASH were renamed as metabolic dysfunction-associated steatotic liver disease (MASLD) and metabolic dysfunction-associated steatohepatitis (MASH), respectively; hence, we introduced the term “high-risk MASH”. Diagnostic values of seven non-invasive models, including FibroScan-aspartate transaminase (FAST), fibrosis-4 (FIB-4), aspartate transaminase to platelet ratio index (APRI), etc. for high-risk MASH have rarely been studied and compared in MASLD.

AIM

To assess the clinical value of seven non-invasive models as alternatives to liver biopsy for diagnosing high-risk MASH.

METHODS

A retrospective analysis was conducted on 309 patients diagnosed with NAFLD via liver biopsy at Beijing Ditan Hospital, between January 2012 and December 2020. After screening for MASLD and the exclusion criteria, 279 patients were included and categorized into high-risk and non-high-risk MASH groups. Utilizing threshold values of each model, sensitivity, specificity, positive predictive value (PPV), and negative predictive values (NPV), were calculated. Receiver operating characteristic curves were constructed to evaluate their diagnostic efficacy based on the area under the curve (AUROC).

RESULTS

MASLD diagnostic criteria were met by 99.4% patients with NAFLD. The MASLD population was analyzed in two cohorts: Overall population (279 patients) and the subgroup (117 patients) who underwent liver transient elastography (FibroScan). In the overall population, FIB-4 showed better diagnostic efficacy and higher PPV, with sensitivity, specificity, PPV, NPV, and AUROC of 26.9%, 95.2%, 73.5%, 72.2%, and 0.75. APRI, Forns index, and aspartate transaminase to alanine transaminase ratio (ARR) showed moderate diagnostic efficacy, whereas S index and gamma-glutamyl transpeptidase to platelet ratio (GPR) were relatively weaker. In the subgroup, FAST had the highest diagnostic efficacy, its sensitivity, specificity, PPV, NPV, and AUROC were 44.2%, 92.3%, 82.1%, 67.4%, and 0.82. The FIB-4 AUROC was 0.76. S index and GPR exhibited almost no diagnostic value for high-risk MASH.

CONCLUSION

FAST and FIB-4 could replace liver biopsy as more effectively diagnostic methods for high-risk MASH compared to APRI, Forns index, ARR, S index, and GPR; FAST is superior to FIB-4.

Steatotic Liver Disease: Pathophysiology and Emerging Pharmacotherapies

Steatotic liver disease (SLD) displays a dynamic and complex disease phenotype. Consequently, the metabolic dysfunction-associated steatotic liver disease (MASLD)/metabolic dysfunction-associated steatohepatitis (MASH) therapeutic pipeline is expanding rapidly and in multiple directions. In parallel, noninvasive tools for diagnosing and monitoring responses to therapeutic interventions are being studied, and clinically feasible findings are being explored as primary outcomes in interventional trials. The realization that distinct subgroups exist under the umbrella of SLD should guide more precise and personalized treatment recommendations and facilitate advancements in pharmacotherapeutics. This review summarizes recent updates of pathophysiology-based nomenclature and outlines both effective pharmacotherapeutics and those in the pipeline for MASLD/MASH, detailing their mode of action and the current status of phase 2 and 3 clinical trials. Of the extensive arsenal of pharmacotherapeutics in the MASLD/MASH pipeline, several have been rejected, whereas other, mainly monotherapy options, have shown only marginal benefits and are now being tested as part of combination therapies, yet others are still in development as monotherapies. Although the Food and Drug Administration (FDA) has recently approved resmetirom, additional therapeutic approaches in development will ideally target MASH and fibrosis while improving cardiometabolic risk factors. Due to the urgent need for the development of novel therapeutic strategies and the potential availability of safety and tolerability data, repurposing existing and approved drugs is an appealing option. Finally, it is essential to highlight that SLD and, by extension, MASLD should be recognized and approached as a systemic disease affecting multiple organs, with the vigorous implementation of interdisciplinary and coordinated action plans. SIGNIFICANCE STATEMENT: Steatotic liver disease (SLD), including metabolic dysfunction-associated steatotic liver disease and metabolic dysfunction-associated steatohepatitis, is the most prevalent chronic liver condition, affecting more than one-fourth of the global population. This review aims to provide the most recent information regarding SLD pathophysiology, diagnosis, and management according to the latest advancements in the guidelines and clinical trials. Collectively, it is hoped that the information provided furthers the understanding of the current state of SLD with direct clinical implications and stimulates research initiatives.

Machine-learning model comprising five clinical indices and liver stiffness measurement can accurately identify MASLD-related liver fibrosis

BACKGROUND & AIMS

aMAP score, as a hepatocellular carcinoma risk score, is proven to be associated with the degree of chronic hepatitis B-related liver fibrosis. We aimed to evaluate the ability of aMAP score for metabolic dysfunction-associated steatotic liver disease (MASLD; formerly NAFLD)-related fibrosis diagnosis and establish a machine-learning (ML) model to improve the diagnostic performance.

METHODS

A total of 946 biopsy-proved MASLD patients from China and the United States were included in the analysis. The aMAP score, demographic/clinical indices and liver stiffness measurement (LSM) were included in seven ML algorithms to build fibrosis diagnostic models in the training set (N = 703). The performance of ML models was evaluated in the external validation set (N = 125).

RESULTS

The AUROCs of aMAP versus fibrosis-4 index (FIB-4) and aspartate aminotransferase-platelet ratio (APRI) in cirrhosis and advanced fibrosis were (0.850 vs. 0.857 [P = 0.734], 0.735 [P = 0.001]) and (0.759 vs. 0.795 [P = 0.027], 0.709 [P = 0.049]). When using dual cut-off values, aMAP had a smaller uncertainty area and higher accuracy (26.9%, 86.6%) than FIB-4 (37.3%, 85.0%) and APRI (59.0%, 77.3%) in cirrhosis diagnosis. The seven ML models performed satisfactorily in most cases. In the validation set, the ML model comprising LSM and 5 indices (including age, sex, platelets, albumin and total bilirubin used in aMAP calculator), built by logistic regression algorithm (called LSM-plus model), exhibited excellent performance. In cirrhosis and advanced fibrosis detection, the LSM-plus model had higher accuracy (96.8%, 91.2%) than LSM alone (86.4%, 67.2%) and Agile score (76.0%, 83.2%), respectively. Additionally, the LSM-plus model also displayed high specificity (cirrhosis: 98.3%; advanced fibrosis: 92.6%) with satisfactory AUROC (0.932, 0.875, respectively) and sensitivity (88.9%, 82.4%, respectively).

CONCLUSIONS

The aMAP score is capable of diagnosing MASLD-related fibrosis. The LSM-plus model could accurately identify MASLD-related cirrhosis and advanced fibrosis.

Degree of Discordance Between FIB-4 and Transient Elastography: An Application of Current Guidelines on General Population Cohort

BACKGROUND & AIMS

In the American Gastroenterological Association/American Association for the Study of Liver Diseases (AGA/AASLD) Clinical Care Pathway, Fibrosis-4 index (FIB-4) is used to stratify patients at risk for metabolic dysfunction-associated steatotic liver disease (MASLD) as low-, indeterminate-, or high-risk for developing advanced liver fibrosis. We assessed the performance of FIB-4 in a general population.

METHODS

Using the 2017 to 2020 National Health and Nutrition Examination Surveys dataset, we selected subjects ≥18 years who had FibroScan data. We followed AGA/AASLD guidelines to identify subjects with characteristics that place them at risk for MASLD-associated liver fibrosis. Other causes of liver disease were excluded. Our final cohort had 3741 subjects. We then categorized these subjects based on recommended FIB-4 cutoffs. FibroScan liver stiffness measurement (LSM) served as the outcome measurement.

RESULTS

Among the 2776 subjects (74.2%) classified as low risk by FIB-4, 277 subjects (10%) were not classified at low risk by LSM, and 75 subjects (2.7%) were classified as high risk by LSM. Among the 86 subjects classified as high risk by FIB-4, 68 subjects (79.1%) were not at high risk by LSM, and 54 subjects (62.8%) were at low risk by LSM. Subjects misclassified by FIB-4 as low risk were older; had a higher body mass index, waist circumference, glycohemoglobin A1c level, alanine transaminase, aspartate transaminase, diastolic blood pressure, controlled attenuation parameter score, white blood cell count, alkaline phosphatase, and fasting glucose level; but had lower high-density lipoprotein, and albumin level (all P < .05). Misclassified subjects were also more likely to have prediabetes/diabetes.

CONCLUSION

Using FIB-4 in the AGA/AASLD guidelines to risk-stratify subjects at risk for MASLD-associated fibrosis results in many subjects being misclassified into the low- and high-risk categories. Therefore, it may be worthwhile considering caution in interpretation and/or alternative strategies.

Algorithms for Early Detection of Silent Liver Fibrosis in the Primary Care Setting

More than one-third of the adult world population has steatotic liver disease (SLD), with a few percent of individuals developing cirrhosis after decades of silent liver fibrosis accumulation. Lack of systematic early detection causes most patients to be diagnosed late, after decompensation, when treatment has limited effect and survival is poor. Unfortunately, no isolated screening test in primary care can sufficiently predict advanced fibrosis from SLD. Recent efforts, therefore, combine several parameters into screening algorithms, to increase diagnostic accuracy. Besides patient selection, for example, by specific characteristics, algorithms include nonpatented or patented blood tests and liver stiffness measurements using elastography-based techniques. Algorithms can be composed as a set of sequential tests, as recommended by most guidelines on primary care pathways. Future use of algorithms that are easy to interpret, cheap, and semiautomatic will improve the management of patients with SLD, to the benefit of global health care systems.

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).

Serum identification of at-risk MASH: The metabolomics-advanced steatohepatitis fibrosis score (MASEF)

BACKGROUND

Early identification of those with NAFLD activity score ≥ 4 and significant fibrosis (≥F2) or at-risk metabolic dysfunction-associated steatohepatitis (MASH) is a priority as these patients are at increased risk for disease progression and may benefit from therapies. We developed and validated a highly specific metabolomics-driven score to identify at-risk MASH.

METHODS

We included derivation (n = 790) and validation (n = 565) cohorts from international tertiary centers. Patients underwent laboratory assessment and liver biopsy for metabolic dysfunction-associated steatotic liver disease. Based on 12 lipids, body mass index, aspartate aminotransferase, and alanine aminotransferase, the MASEF score was developed to identify at-risk MASH and compared to the FibroScan-AST (FAST) score. We further compared the performance of a FIB-4 + MASEF algorithm to that of FIB-4 + liver stiffness measurements (LSM) by vibration-controlled transient elastography (VCTE).

RESULTS

The diagnostic performance of the MASEF score showed an area under the receiver-operating characteristic curve, sensitivity, specificity, and positive and negative predictive values of 0.76 (95% CI 0.72-0.79), 0.69, 0.74, 0.53, and 0.85 in the derivation cohort, and 0.79 (95% CI 0.75-0.83), 0.78, 0.65, 0.48, and 0.88 in the validation cohort, while FibroScan-AST performance in the validation cohort was 0.74 (95% CI 0.68-0.79; p = 0.064), 0.58, 0.79, 0.67, and 0.73, respectively. FIB-4+MASEF showed similar overall performance compared with FIB-4 + LSM by VCTE ( p = 0.69) to identify at-risk MASH.

CONCLUSION

MASEF is a promising diagnostic tool for the assessment of at-risk MASH. It could be used alternatively to LSM by VCTE in the algorithm that is currently recommended by several guidance publications.

Hepatocyte Deubiquitinating Enzyme OTUD5 Deficiency is a Key Aggravator for Metabolic Dysfunction-Associated Steatohepatitis by Disturbing Mitochondrial Homeostasis

BACKGROUND & AIMS

Metabolic dysfunction-associated steatohepatitis (MASH) is a common chronic liver disease worldwide. No effective pharmacologic therapies for MASH have been developed; to develop such promising drugs, the underlying mechanisms regulating MASH need to be elucidated. Here, we aimed to determine the role of ovarian tumor domain-containing protein 5 (OTUD5) in MASH progression and identify a specific mechanism.

METHODS

The expression levels of OTUD subfamily under palmitic acid/oleic acid (PAOA) stimulation were screened. OTUD5 expression was assessed in human liver tissues without steatosis, those with simple steatosis, and those with MASH. MASH models were developed in hepatocyte-specific Otud5-knockout mice that were fed high-fat high-cholesterol and high-fat high-cholesterol plus high-fructose/sucrose diet for 16 weeks.

RESULTS

The expression of OTUD5 was down-regulated in fatty liver and was negatively related to the progression of MASH. Lipid accumulation and inflammation were exacerbated by Otud5 knockdown but attenuated by Otud5 overexpression under PAOA treatment. Hepatocyte-specific Otud5 deletion markedly exacerbated steatosis, inflammation, and fibrosis in the livers of 2 MASH mouse models. We identified voltage-dependent anion channel 2 (VDAC2) as an OTUD5-interacting partner; OTUD5 cleaved the K48-linked polyubiquitin chains from VDAC2, and it inhibited subsequent proteasomal degradation. The anabolic effects of OTUD5 knockdown on PAOA-induced lipid accumulation were effectively reversed by VDAC2 overexpression in primary hepatocytes. Metabolomic results revealed that VDAC2 is required for OTUD5-mediated protection against hepatic steatosis by maintaining mitochondrial function.

CONCLUSIONS

OTUD5 may ameliorate MASH progression via VDAC2-maintained mitochondrial homeostasis. Targeting OTUD5 may be a viable MASH-treatment strategy.

From NAFLD to HCC: Advances in noninvasive diagnosis

Non-alcoholic fatty liver disease (NAFLD) has gradually become one of the major liver health problems in the world. The dynamic course of the disease goes through steatosis, inflammation, fibrosis, and carcinoma. Before progressing to carcinoma, timely and effective intervention will make the condition better, which highlights the importance of early diagnosis. With the further study of the biological mechanism in the pathogenesis and progression of NAFLD, some potential biomarkers have been discovered, and the possibility of their clinical application is gradually being discussed. At the same time, the progress of imaging technology and the emergence of new materials and methods also provide more possibilities for the diagnosis of NAFLD. This article reviews the diagnostic markers and advanced diagnostic methods of NAFLD in recent years.

Practical diagnosis of cirrhosis in non-alcoholic fatty liver disease using currently available non-invasive fibrosis tests

Unlike for advanced liver fibrosis, the practical rules for the early non-invasive diagnosis of cirrhosis in NAFLD remain not well defined. Here, we report the derivation and validation of a stepwise diagnostic algorithm in 1568 patients with NAFLD and liver biopsy coming from four independent cohorts. The study algorithm, using first the elastography-based tests Agile3+ and Agile4 and then the specialized blood tests FibroMeterV3G and CirrhoMeterV3G, provides stratification in four groups, the last of which is enriched in cirrhosis (71% prevalence in the validation set). A risk prediction chart is also derived to allow estimation of the individual probability of cirrhosis. The predicted risk shows excellent calibration in the validation set, and mean difference with perfect prediction is only -2.9%. These tools improve the personalized non-invasive diagnosis of cirrhosis in NAFLD.

Exploring Opportunities to Enhance the Screening and Surveillance of Hepatocellular Carcinoma in Non-Alcoholic Fatty Liver Disease (NAFLD) through Risk Stratification Algorithms Incorporating Ultrasound Elastography

Non-alcoholic fatty liver disease (NAFLD), with its progressive form, non-alcoholic steatohepatitis (NASH), has emerged as a significant public health concern, affecting over 30% of the global population. Hepatocellular carcinoma (HCC), a complication associated with both cirrhotic and non-cirrhotic NAFLD, has shown a significant increase in incidence. A substantial proportion of NAFLD-related HCC occurs in non-cirrhotic livers, highlighting the need for improved risk stratification and surveillance strategies. This comprehensive review explores the potential role of liver ultrasound elastography as a risk assessment tool for HCC development in NAFLD and highlights the importance of effective screening tools for early, cost-effective detection and improved management of NAFLD-related HCC. The integration of non-invasive tools and algorithms into risk stratification strategies could have the capacity to enhance NAFLD-related HCC screening and surveillance effectiveness. Alongside exploring the potential advancement of non-invasive tools and algorithms for effectively stratifying HCC risk in NAFLD, we offer essential perspectives that could enable readers to improve the personalized assessment of NAFLD-related HCC risk through a more methodical screening approach.

Artificial Intelligence Applications in Hepatology

Over the past 2 decades, the field of hepatology has witnessed major developments in diagnostic tools, prognostic models, and treatment options making it one of the most complex medical subspecialties. Through artificial intelligence (AI) and machine learning, computers are now able to learn from complex and diverse clinical datasets to solve real-world medical problems with performance that surpasses that of physicians in certain areas. AI algorithms are currently being implemented in liver imaging, interpretation of liver histopathology, noninvasive tests, prediction models, and more. In this review, we provide a summary of the state of AI in hepatology and discuss current challenges for large-scale implementation including some ethical aspects. We emphasize to the readers that most AI-based algorithms that are discussed in this review are still considered in early development and their utility and impact on patient outcomes still need to be assessed in future large-scale and inclusive studies. Our vision is that the use of AI in hepatology will enhance physician performance, decrease the burden and time spent on documentation, and reestablish the personalized patient-physician relationship that is of utmost importance for obtaining good outcomes.