MetALD: Clinical aspects, pathophysiology and treatment

Summary

Metabolic dysfunction-associated steatotic liver disease (MASLD) and alcohol-related liver disease (ALD) are the most prevalent causes of chronic liver disease worldwide. Both conditions have many pathophysiological mechanisms in common, such as altered lipid and bile acid metabolism, and share some similar clinical features. Furthermore, metabolic risk factors and alcohol often co-exist in the same individuals and have recently been shown to act synergistically to markedly increase the risk of liver disease. Given the high prevalence and impact of this interaction, steatotic liver disease due to the combination of metabolic dysfunction and moderate-to-high alcohol intake has been termed MetALD in the new steatotic liver disease nomenclature, attracting the interest of the scientific community. Subsequent studies have investigated the prevalence of MetALD, which ranges from 1.7% to 17% in cohorts of patients with steatotic liver disease, depending on the population setting and study design. A few cohort studies have also assessed the prognosis of this patient population, with preliminary data suggesting that MetALD is associated with an intermediate risk of liver fibrosis, decompensation and mortality among steatotic liver disease subtypes. In this review article, we examine the clinical evidence and the experimental models of MetALD and discuss the clinical implications of the term for early detection and management. We provide insight into the pathophysiological mechanisms of the synergistic effect of alcohol and metabolic risk factors, possible screening strategies, the use of biomarkers and emerging models of care, as well as potential therapeutic interventions with a special focus on medications for MASLD, highlighting the most promising drugs for patients with MetALD.

Graphical abstract

Created in BioRender. Pose, E. (2024) https://BioRender.com/s75w153

Keypoints.

• •The term metabolic dysfunction and alcohol-related liver disease (MetALD) was recently adopted to refer to patients with steatotic liver disease (SLD) due to at least one metabolic risk factor and moderate-to-high alcohol consumption.
• •The prevalence of MetALD ranges from 1.7% to 17%, depending on the setting, and preliminary data suggest that, among SLD subtypes, it is associated with an intermediate risk of liver fibrosis, decompensation and death.
• •Targeted screening and the sequential use of biomarkers of liver fibrosis are probably the most cost-effective strategies for early detection.
• •Most medications investigated for metabolic dysfunction-associated steatotic liver disease (MASLD) might also be effective in MetALD. Glucagon-like peptide-1 receptor agonists and fibroblast growth factor 21 analogues are the most promising drugs for MetALD due to their potential effect on reducing alcohol intake.

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) and alcohol-related liver disease (ALD) are the most prevalent causes of chronic liver disease worldwide and account for most liver-related hospitalizations and deaths.¹ The prevalence of MASLD and ALD has been increasing in recent decades due to the rise in the prevalence of risk factors such as obesity and type 2 diabetes mellitus, along with harmful alcohol use.^2,3

Both MASLD and ALD share a similar clinical course, from asymptomatic compensated stages to advanced stages of decompensated cirrhosis with altered liver synthetic function, and are associated with similar histological findings in the liver. Moreover, both entities even share many pathophysiological mechanisms of liver injury. Despite their long-known similarities, MASLD and ALD have traditionally been considered two separate conditions. However, several classical^4,5 as well as more recent studies have shown that risk factors of both conditions frequently co-exist and have a synergistic effect in driving liver disease development and progression.^6,7 This synergistic effect even occurs when consuming 1.5 drinks per day for females and 3 for males,⁸ which corresponds to the World Health Organization threshold of moderate risk alcohol consumption.⁹ Thus, the combination of metabolic syndrome and moderate alcohol consumption predisposes to the development of advanced liver disease.

Given the aforementioned considerations and the potential stigma related to the term “fatty” present in the former definition (NAFLD, “nonalcoholic fatty liver disease”), a Delphi-based consensus study with numerous stakeholders including liver experts and patient advocates supported by the European Association for the Study of the Liver, the American Association for the Study of Liver Diseases and the Asociación Latinoamericana para el Estudio del Hígado, proposed a revised nomenclature to describe patients with steatosis associated with metabolic dysfunction, alcohol consumption or both.¹⁰ In this new nomenclature, the term steatotic liver disease (SLD) was chosen to encompass the different causes of liver disease characterised by fat droplet accumulation as the common mechanism of liver injury. Patients with SLD were further subclassified into four groups: 1) MASLD (replacing NAFLD), including patients consuming less than 2 drinks per day for females and 3 for males; 2) ALD, including patients consuming more than 5 drinks per day for females and 6 for males; 3) metabolic dysfunction and alcohol-related steatotic liver disease (MetALD), which includes patients with at least 1 metabolic risk factor (MRF) and alcohol intake ranging from 2 to 5 drinks per day for females and from 3 to 6 for males; and 4) others (i.e., specific drug-induced liver injury causing steatosis). Importantly, the introduction of the term MetALD allowed for a better description of patients who had been largely overlooked, highlighting the synergism between MRFs and even moderate alcohol consumption.

Considering the emerging evidence regarding the synergistic adverse impact of metabolic dysfunction and alcohol consumption, in this review article we aim to summarise the current knowledge regarding clinical and pathophysiological aspects of SLD, focusing on the implications of the new term MetALD, both in clinical practice and research.

Clinical aspects

Since the introduction of the new term MetALD, most published studies on this topic have assessed the prevalence of this condition in cohorts of patients previously classified with ALD or MASLD[11], [12], [13], [14] or in large datasets such as the United Kingdom biobank¹⁵ (Table 1). The prevalence of MetALD in these studies ranged from 1.7% to 17.0%, mostly depending on the population setting and the definition of SLD used. Studies performed in the general population have reported the lowest prevalence, ranging from 1.7% to 4.5%, while the only study based on the primary care setting found a prevalence of 5.6%.¹⁶ A Danish study performed in an alcohol use-enriched cohort, including patients from the general population and from secondary care, reported the highest prevalence at 17.0%.¹⁷ The definition of SLD used in these studies was highly heterogeneous: four studies used controlled attenuation parameter (CAP) by transient elastography (TE) with different thresholds for steatosis, two used MRI-proton density fat fraction (MRI-PDFF), one used abdominal ultrasound and one used the fatty liver index. As expected, studies using more sensitive techniques, such as MRI-PDFF or CAP with lower thresholds, reported the highest prevalence. However, it should be noted that comprehensive data on key variables that are needed for the diagnosis of MetALD are lacking in these studies. Moreover, several studies derive from the same nationwide cohort;[11], [12], [13] thus, the findings are quite similar.

Table 1.

Studies assessing the prevalence and/or clinical characteristics of patients with MetALD using the new consensus definition of SLD.

Reference	Population	Study design	MetALD prevalence	Association with fibrosis	Association with clinical outcomes
Schneider CV et al., 202415	Adults >36 years of age from the UK Biobank who underwent an MRI-PDFF	Cross-sectional study Steatosis based on MRI-PDFF	In the total cohort: 2.1% In the SLD subgroup: 7.9%	—	—
Choe HJ et al., 202416	Adults >35 years of age from the general Korean population	Prospective cohort study SLD based on Fatty Liver Index ≥30	In the total cohort: 4.5% In the SLD subgroup: 9.9%	Significant fibrosis (FIB-4 ≥1.3 in individuals aged 35-64 years; FIB-4 ≥2.0 in individuals aged 65 years or older): MASLD: 27% MetALD: 38% ALD: 41%	Cardiovascular events: HR of 1.27 (95% CI 1.12-1.45) for MASLD, 1.88 (1.33-2.65) for MetALD and 1.94 (1.01-3.77) for ALD. Notably risk was higher in the MetALD than in the MASLD group (p <0.05) Reference: patients without SLD
Israelsen M et al., 202417	Adults with current or previous excessive alcohol intake with no history of decompensated cirrhosis recruited from hospital liver clinics, municipal alcohol rehabilitation centers and public health portal in Denmark	Prospective cohort study SLD based on CAP ≥290 dB/m or steatosis on liver biopsy examination	In the total cohort: 17% In the SLD subgroup: 24%	Significant fibrosis (METAVIR F2 or above): MASLD: 69% MetALD: 53% ALD: 47%	Risk of decompensation: HR of 4.73 (95% CI 1.03-21.6) for MASLD, 7.69 (1.66-35.6) for MetALD and 10.2 (2.24-46.4) for ALD Mortality: HR of 2.30 (1.08-4.90) for MASLD, 2.94 (1.31-6.58) for MetALD and 3.57 (1.64-7.80) for ALD Reference: patients without SLD
Lee BP et al., 202412	Adults from the general United States population (NHANES cohort) with complete transient elastography examination	Cross-sectional study SLD based on CAP values	In the total cohort: 2.0% In the SLD subgroup: 5.8%	Increasing amount of alcohol consumption below the threshold of MetALD were associated with advanced liver fibrosis in male individuals with SLD, but not in female individuals	—
Miwa T et al., 202418	Staff and faculty members undergoing annual occupational checkup at Gifu University (Japan) with complete clinical, laboratory and ultrasound data	Cross-sectional study SLD based on ultrasound findings	In the total cohort: 1.7% In the SLD subgroup: 2.5%	Individuals with MetALD had significantly higher median FIB-4 index values than individuals with MASLD (0.92 [0.81-1.17] vs. 0.69 [0.53-0.87], respectively; p <0.05), but similar to individuals with ALD (0.92 [0.81-1.17] vs. 1.09 [0.65-1.24], respectively; p >0.05)	—
Kalligeros M et al., 202311	Adults from the general United States population (NHANES cohort) with complete transient elastography examination	Cross-sectional study SLD based on CAP ≥285 dB/m	In the total cohort: 2.6% In the SLD subgroup: 6.8%	Significant fibrosis (LSM >8.6 kPa): MASLD: 21% MetALD: 13% ALD: 12%	—
Ciardullo S et al., 202313	Adults from the general United States population (NHANES cohort) with complete transient elastography examination	Cross-sectional study Steatosis based on CAP ≥274 dB/m	In the total cohort: 3.2% In the SLD subgroup: 7.7%	Significant fibrosis (LSM >8 kPa): MASLD: 15% MetALD: 10% ALD: 0%a	—
Lee CM et al., 202314	Adults attending primary care checkups in Korea who underwent MRE and MRI-PDFF	Retrospective cohort study Steatosis based on MRI-PDFF	In the total cohort: 5.6% In the SLD subgroup: 14.4%	The MetALD subgroup had significantly higher mean values of MRE than the MASLD and ALD subgroups (2.30±0.56 vs. 2.22±0.64 vs. 2.3±0.56 kPa, respectively)	—

ALD, alcohol-related liver disease; CAP, controlled attenuation parameter; FIB-4, fibrosis-4; HR, hazard ratio; LSM, liver stiffness measurement; MASLD, metabolic dysfunction-associated steatotic liver disease; MetALD, metabolic dysfunction and alcohol-related steatotic liver disease; METAVIR, meta-analysis of histological data in viral hepatitis; MRE, magnetic resonance elastography; MRI-PDFF, MRI-proton density fat fraction; NHANES, national health and nutrition examination survey; SLD, steatotic liver disease.

^aOnly 5 patients in the ALD group.

Regarding the association of MetALD with the presence of liver fibrosis, two Asian studies reported higher values of FIB-4 in patients with MetALD compared to patients with MASLD, but lower values compared to patients with ALD.^16,18 Interestingly, one of these studies¹⁶ also reported a higher risk of cardiovascular events in the MetALD group compared to the MASLD group. In another Korean study in which fibrosis was assessed by magnetic resonance elastography, patients with MetALD had significantly higher liver stiffness values than patients with MASLD or ALD.¹⁴

Finally, emerging studies from Northern Europe and North America strongly suggest that patients with MetALD have a higher risk of liver decompensation and mortality than patients meeting MASLD criteria,^19,20 but a lower risk than patients with ALD.^17,21 Regarding hepatocellular carcinoma (HCC) development, two studies from Asia have reported a stepwise increase in risk from MASLD to MetALD and then ALD.^22,23 These preliminary data suggest that MetALD is associated with an intermediate risk of liver disease severity among SLD subtypes. In terms of cardiovascular risk, information is scarce and controversial; one study has shown increased risk in patients with MetALD compared to patients with MASLD,¹⁶ while two other studies reported inverse results.^21,24 Future studies should better define the clinical course of MetALD and the factors that influence patient outcomes.

Effects of alcohol on patients with MASLD

Several studies have investigated the effect of moderate alcohol consumption in patients with MASLD.^6,8,[25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41] Most of them were retrospective studies based on large databases including population-based cohorts with thousands of individuals, in which individuals with steatosis due to MRFs were investigated. MASLD was diagnosed based on the presence of steatosis assessed by different methods (liver biopsy, ultrasound, CAP, computed tomography, hepatic steatosis index or diagnostic coding). Information on alcohol consumption history was collected heterogeneously and by self-reporting. Therefore, the quality of the data was suboptimal and prospective studies including alcohol biomarkers are needed. The outcomes included in these studies were: presence, risk of development, improvement or worsening of liver steatosis,^35,39 liver fibrosis,^29,31,33,36 steatohepatitis,⁴⁰ risk of HCC development^26,32 and all-cause mortality.^6,25,27,41 The majority of these studies found a negative impact of alcohol consumption on liver disease even among consumers of low or moderate amounts.^{6,25,26,29,30,32,36} Importantly, the presence of diabetes markedly increases the risk of significant liver disease.^42,43

An issue deserving some discussion is the impact of binge drinking in patients with MASLD. Binge drinking is defined as consuming ≥4 drinks for females or ≥5 drinks for males on the same occasion on at least 1 day in the past 30 days.⁴⁴ While several studies have shown deleterious effects of binge drinking on preclinical models of MASLD,^45,46 clinical data on this topic are limited. The highest quality evidence derives from a Finnish population-based study that showed supra-additive interactions between binge drinking and metabolic syndrome on the risk of liver-related events, independently of average alcohol consumption.⁴⁷ The new SLD nomenclature uses weekly or average daily alcohol intake to subclassify patients and, therefore, people with a binge drinking pattern might be incorrectly classified.

Another controversial topic in the past years was whether small amounts of alcohol, especially wine, could have beneficial effects in patients with MASLD. This hypothesis was based on data deriving from some epidemiological studies in large populations showing a beneficial association of moderate alcohol consumption with both liver- and non-liver-related outcomes among patients with MASLD.^31,[33], [34], [35]^{,37,39,40,48,49} However, as previously discussed, there is an increasing body of evidence showing that even small amounts of alcohol adversely affect the development and progression of liver disease. Therefore, the protective hypothesis is highly questionable and is not accepted by the hepatology community.

In conclusion, based on the biological plausibility of the negative impact of the combination of MRFs and alcohol in the liver, and the fact that the majority of studies show a negative impact of alcohol use on the risk of developing liver fibrosis and relevant clinical outcomes, we conclude that even moderate amounts of alcohol have a negative effect on the natural history of MASLD. Prospective studies following the new definition of SLD and considering the duration and pattern of alcohol use are needed to better address this issue.

Effects of metabolic syndrome on patients with ALD

The amount of evidence regarding the effect of MRFs on patients with ALD is scarce. One of the first studies in the field included a large cohort of more than 1,000 people with hazardous drinking assessed by liver biopsy and classified according to the histological diagnosis. The authors found obesity to be a risk factor for the presence of steatosis, fibrosis and cirrhosis.⁴ Regarding the effect of diabetes per se on ALD, a recent study found that the homeostatic model assessment for insulin resistance index increased in a stepwise manner in progressive stages of ALD, being the strongest predictor of the presence of liver fibrosis and inflammation, with a hazard ratio greater than x3 for liver fibrosis.⁵⁰ Similarly, a population-based study including more than 3,000 individuals screened for liver fibrosis found that among people with hazardous drinking, the presence of metabolic syndrome increased the risk of liver fibrosis 3.7-fold.⁷

In a registry-based study that merged two cohorts from Scotland, with approximately 4,000 individuals, there was an increased risk of liver disease in individuals with either hazardous alcohol consumption or obesity. Interestingly, it showed a synergistic effect of both cofactors, with a 2.89-fold increase in the risk of liver disease compared to the expected effect from the addition of both.⁵ Further studies investigating the synergistic vs. additive interaction between alcohol and MRFs in the newly defined population with MetALD will address this and other relevant unanswered questions.

Causes of death and comorbidities

One of the aspects that has generated more debate regarding MetALD is the relative contribution of each aetiological factor in the progression of liver disease and patient outcomes. To date, most studies have investigated the long-term prognosis and causes of death either in patients with ALD or MASLD.

Several recent registry-based studies including a large number of individuals have tried to compare the natural history of MASLD and ALD.[51], [52], [53], [54], [55], [56] Globally, these studies found that the risk of liver-related and non-liver-related deaths increased with the progression of liver disease both in individuals with MASLD and ALD. Regarding the studies in patients with ALD, two recent studies based on European cohorts, accounting altogether for more than 26,000 individuals, found that liver-related mortality remained as the main cause of death in this population, even at early stages of liver disease. Interestingly, these studies also found an increase in non-liver-related causes of death among individuals with ALD, mainly cardiovascular diseases and extrahepatic cancer.^54,56 However, it should be noted that ALD mortality, especially non-liver-related mortality, has been decreasing in recent decades⁵⁷ possibly due to improvements in the management of chronic conditions and cancer. In the case of MASLD, a study based on the NHANES-III (Third National Health and Nutrition Examination Survey) including individuals with MASLD at early stages with a median follow-up of 8 years found that liver diseases were the third most common cause of death in this population, behind cardiovascular diseases and malignancies.⁵⁸ In contrast, in several studies including patients with advanced fibrosis due to MASLD the main cause of death was liver disease.^59,60 Specific studies assessing the long-term outcome and prognostic drivers of patients that meet MetALD criteria are very limited and are anticipated in the coming years.

Pathophysiology of the synergistic effect of alcohol and metabolic syndrome

Common mechanisms

Pathogenic mechanisms underlying MASLD and ALD have usually been investigated independently. However, it is well known that both diseases share important histological features and pathophysiological processes responsible for steatosis, fibrosis, chronic inflammation, and hepatocellular dysfunction. Importantly, both diseases share the genetic factors influencing disease progression.⁶¹ Therefore, a combination of common mechanisms involved in MASLD and ALD pathogenesis are likely to play a role in MetALD (Fig. 1).

Fig. 1.

Pathogenic mechanisms underlying MetALD.

A combination of common mechanisms underlying MASLD and ALD may be considered the hallmarks of MetALD. Created in BioRender. Pose, E. (2024) https://BioRender.com/d99w868. ALD, alcohol-related liver disease; AUD, alcohol use disorder; DAMPs, damage-associated molecular patterns; ER, endoplasmic reticulum; FXR, farnesoid X receptor; MASLD, metabolic dysfunction-associated steatotic liver disease; MetALD, metabolic dysfunction and alcohol-related liver disease; TGR5, Takeda G-protein coupled receptor 5.

Altered lipid metabolism

Steatosis in hepatocytes is an early pathological feature found in both MASLD and ALD.⁶² Lipid droplet formation and accumulation is a consequence of an imbalance between lipid uptake, synthesis, degradation, and export of free fatty acids, which is mainly associated with insulin resistance.⁶³ In MASLD, increased adipose tissue and insulin resistance lead to higher levels of free fatty acids in circulation, contributing to de novo lipogenesis and hepatic lipid accumulation.⁶⁴ Similarly, chronic alcohol consumption has been associated with adipose tissue lipolysis, upregulation of hepatic fatty acid transporters and enhanced activity of key lipogenic regulators.⁶⁵ In parallel, decreased fatty acid oxidation and disturbed very-low-density lipoprotein secretion has been described in both conditions, fueling hepatic steatosis.⁶⁶

Immune response

Both innate and adaptive immune cells are implicated in the progression of MASLD and ALD.^67,68 In both disease conditions, macrophages can respond to multiple stimuli, being able to polarise to different phenotypes.^69,70 During recent years, single-cell RNA sequencing has enabled the discovery of several populations of macrophages in MASLD, underscoring a multifaceted role of these cells depending on the disease stage and their localisation.^70,71 However, further studies are required to confirm such heterogeneity in ALD, where macrophage characterisation has been more broadly addressed. Moreover, neutrophils are recruited to the liver in patients with MASLD and ALD, where they acquire an activated phenotype contributing to pathogenic processes and inflammation through the release of reactive oxygen species (ROS), neutrophil extracellular traps, cytokines and proteases.⁷² Regarding adaptive immunity, altered features in T cells and unconventional populations have been found in the settings of MASLD and ALD, amplifying liver inflammatory response.^73,74

Hepatic stellate cell activation

Hepatic stellate cells (HSCs) are activated and play a key role in fibrosis progression in both MASLD and ALD. Activated HSCs become proliferative and contractile and are responsible for the excessive extracellular matrix deposition and reduced matrix degradation that lead to fibrosis development.^75,76 Moreover, activated HSCs secrete a wide range of pro-inflammatory cytokines such as the C–C motif chemokine ligand 2 and IL8, contributing to inflammatory cell recruitment and chronic inflammation.⁷⁷

Hepatocellular damage and death

Hepatocellular injury and death are common findings in steatohepatitis in both MASLD and ALD. Also, hepatocyte injury contributes to liver inflammation and fibrosis, enhancing disease progression. However, the mechanisms of cell death in MASLD and ALD are not identical. In the setting of MASLD, animal models have shown that toxic lipids such as ceramides accumulate within hepatocytes promoting cell damage through endoplasmic reticulum stress, altered mitochondrial activity, lysosomal dysfunction, and inflammasome activation; eventually inducing cell death pathways.^78,79 Following chronic alcohol exposure, increased ROS production and endoplasmic reticulum stress have been highlighted as major triggers of hepatocyte injury and activation of cell death pathways.⁸⁰ As a result of sublethal cell damage, hepatocytes release inflammatory and pro-fibrogenic mediators, including cytokines and damage-associated molecular patterns.⁷⁸ In both disease conditions, apoptosis has been characterised in mice and cell line models as the predominant mechanism of hepatocyte cell death, with impaired autophagy being a potential driver of this process.⁸¹ However, other forms of cell death including ferroptosis, pyroptosis and necroptosis have been presented as simultaneous mechanisms of parenchymal dysfunction contributing to fibrosis and inflammation.⁸²

Intestinal dysfunction

Dysbiosis is a common hallmark in both patients with ALD and MASLD, although different specific microbial alterations have been found for each aetiology.⁸³ Similar to the toxic effect of chronic alcohol consumption in patients with ALD, the altered microbiome in MASLD results in bacterial overgrowth and endogenous alcohol production.^84,85 Chronic exposure to high levels of ethanol results in intestinal permeability and increased concentrations of circulating endotoxins such as lipopolysaccharide.⁸⁶ Liver immune cells, among which Kupffer cells play a major role, sense these endotoxins via Toll-like receptors, amplifying the inflammatory response and contributing to fibrosis and steatosis by releasing ROS and pro-inflammatory cytokines, including tumour necrosis factor-α and IL-6.

Disrupted bile acid metabolism

Besides their role in regulating fat digestion and absorption in the small intestine, bile acids are a family of key signalling molecules, acting with different affinities and functions as natural ligands for several nuclear and membrane receptors, including farnesoid X receptor (FXR) and Takeda G-protein coupled receptor 5 (TGR5).⁸⁷ Among other effects, FXR activation regulates glucose homeostasis and inhibits de novo lipogenesis, thus attenuating hepatic steatosis.⁸⁸ In addition, TGR5 signalling has been found to improve insulin sensitivity showing a potential anti-steatotic effect.^89,90 FXR activation by primary chenodeoxycholic acid contrasts with the inhibitory effect of tauro α- and tauro β-muricholic acids.⁹¹ On the other hand, TGR5 is activated by chenodeoxycholic acid and secondary lithocholic and deoxycholic acids.⁹² Though studies on human samples are inconclusive, bile acid pool size and composition have been shown to be altered in relevant mouse models of steatohepatitis,^93,94 being proposed as a potential therapeutic mechanism in both MASLD and ALD.

Ductular reaction

In advanced stages of both MASLD and ALD, a ductular reaction deriving from both metaplastic hepatocytes and bile ducts occurs, leading to the accumulation of bile duct-like structures (typically keratin 7-positive) that correlate with poor prognosis.^95,96 Studies based on human samples and animal models have shown that ductular reaction contributes to hepatic new-vessel formation, inflammation and periportal immune cell recruitment through the release of inflammatory and angiogenic mediators.⁹⁷ Furthermore, several studies have associated ductular reaction cells with HSC activation in mouse models of liver injury.^98,99

Experimental studies on synergism between alcohol and metabolic risk factors

The additive/synergistic effects of metabolic dysfunction and alcohol intake have been investigated using preclinical models of MetALD (Table 2). Such models have reproduced human disease hallmarks such as obesity, steatosis, hepatomegaly, cell death, oxidative stress, fibrosis, and inflammation. Moreover, long-term dual feeding resembled human disease progression, leading to extensive hepatocellular injury, exacerbated histological disease features and development of HCC.¹⁰⁰ Accordingly, other animal models combining diet-induced steatosis with acute alcohol binge or chronic ethanol administration have been able to elucidate potential drivers of steatohepatitis and fibrosis, such as C-X-C motif chemokine ligand 2-dependent neutrophil recruitment or macrophage M1 polarisation.^46,[101], [102], [103], [104] Other studies using genetically modified animal models with spontaneous obesity have reasoned that ethanol binge superimposed to primary steatosis is an essential second hit to model human disease progression, demonstrating that occasional alcohol intake is a risk factor for increased pathogenesis.[105], [106], [107] Recently, steatohepatitis was also modelled in non-obese mice due to the absence of obesity in a representative proportion of the MASLD population, thus underscoring the need for non-classical mouse models lacking metabolic syndrome but still presenting with features of steatohepatitis.¹⁰³ In contrast, few in vitro and ex vivo studies exploring synergistic mechanisms have been performed and almost all of them used animal-derived cells and tissues, thus limiting translational relevance.[108], [109], [110] Of note, precision-cut liver slices from human liver tissue efficiently modelled disease pathogenic features such as lipid synthesis, hepatotoxicity, inflammation and fibrogenesis upon stimulation with ethanol, fatty acids and lipopolysaccharide.¹¹¹

Table 2.

Preclinical models of MetALD investigating the synergistic effects of metabolic dysfunction and alcohol consumption.

Reference	Model	Species	Methods	Duration	Systemic/adipose tissue (AT) outcomes	Hepatic outcomes
Benedé-Ubieto R et al., 2021100	DUAL model	C57BL6/J mice	WD plus 10% vol/vol EtOH absolute in sweetened drinking water containing 6.75% D-glucose	23 weeks and 52 weeks of feeding	↑Body weight ↑BMI ↑ALT, AST and LDH ↑Cholesterol levels ↑Adipocyte size Macrophage crown-like structures (AT) ↑F4/80+ and CD45+ cells (AT) ↑Fibrosis (AT)	↑Fibrosis ↑Steatosis ↑Oxidative and ER stress ↑Cell death ↑F4/80+, CD45+ cells and neutrophils ↑TNFa expression ↑HSCs activation and fibrosis HCC development (after 52 weeks)
Minato T et al., 2014105	“Two-hit” model	OLETF rats	Spontaneous obesity in 30–33-week-old OLETF rats + 10 ml of 10% ethanol by gavage for 5, 3, 2 and 1 day/week	3 weeks of ethanol treatment	↑Serum glucose ↑ALT, AST and LDH ↑Serum TG ↓Serum adiponectin	↑Steatosis ↑Lobular inflammation ↑Cell death ↑Hepatocyte ballooning ↑NAS ↑TNFα expression ↑Oxidative stress
Carmiel-Haggai M et al., 2003107	Genetic obesity + acute ethanol binge	fa/fa Zucker rats	Spontaneous genetic obesity in 15-week-old fa/fa Zucker rats + 35% ethanol (4 g/kg) by gavage every 12 h	3 days of ethanol treatment	↑ALT	↑Steatosis ↑Inflammation (zone 3) ↑Hepatocyte ballooning ↓Antioxidant defence (GSH, catalase, GPX, GR, SOD) ↑Oxidative stress and lipid peroxidation ↑Hepatocyte apoptosis
Chang B et al., 201546 Zhou Z et al., 2018101	HFD + single ethanol binge	C57BL/6J mice	HFD (60% of kcal as fat) + single dose of ethanol (5 g/kg) by gavage the last day of HFD feeding	12 weeks of feeding	↑Serum TG ↑ALT and AST ↑Inflammatory gene CXCL11 (AT)	↑Chemokine (CXCL12) levels and neutrophil recruitment ↑F4/80+ cells ↑HSCs activation and fibrosis ↑Oxidative stress and lipid peroxidation ↓PPARγ levels
Shiraishi S et al., 2024103	Non-obese steatohepatitis + chronic ethanol	C57BL/6J mice	STHD-01 (40% fat and 5% cholesterol) + 5 g/dl ethanol (first week) and 10 g/dl ethanol in water (the rest of the weeks)	6 or 12 weeks of feeding	↓Body weight ↓Serum glucose ↑Cardiac dysfunction	↑Liver weight ↑CD68+ cells and pro-inflammatory cytokine levels ↑Fibrosis
Xu J et al., 2011102	Moderate obese mice + low/high ethanol chronic infusion	C57BL/6J mice	Intragastric high-fat liquid diet (37% cal. from corn oi) + intragastric low-dose (23 g/kg/day) or high-dose (32 g/kg/day) of ethanol	7 weeks of feeding	↑Body weight ↑ALT ↑M2 Arg1+ macrophages (AT)	↑Liver weight ↑Steatosis and liver TG ↑Pericellular fibrosis ↑Inflammatory foci ↑ER stress ↓Mitochondrial biogenesis, ATP, and fatty acid oxidation ↑M1 iNOS+ macrophages
Everitt H et al., 2013106	ob/ob mice + PUFA diet with chronic ethanol	ob/ob C57BL/6J mice	ob/ob mice fed with PUFA liquid diet + increasing ethanol doses until accounting for 27.5% of total diet calories	4 weeks of feeding	↑Tnfa mRNA levels ↑ALT, AST	↑Liver/body weight ratio ↑Liver TG and cholesterol ↑Steatosis ↓Sirt1 signalling Altered lipin-1 location within the cell
Song M et al., 2016104	High-fructose diet + chronic ethanol	C57BL/6J mice	High-fructose diet (60% v/v) + 20% (v/v) ethanol ad libitum from the 9th week through the 18th week	18 weeks of feeding	↑ALT, AST	↑Liver/body weight ratio ↑Liver weight ↑Microvesicular steatosis ↑M1 macrophage polarisation ↑Inflammation
Rastovic U et al., 2023111	Human PCLS + ethanol + FA + LPS	Tumour-free distal portion of human liver	PCLS stimulated with 250 mM ethanol + 0.1 mM oleic/linoleic acids + 10 μg/ml LPS	1, 3 or 5 days of stimulation	↑Supernatant levels of pro-inflammatory cytokines (TNFα, IL-6, IL-8, IL1b)	↑Hepatocyte death
Duryee MJ et al., 2014110	Obese rat PCLS + ethanol	Liver tissue from obese Sprague-Dawley rats	PCLS from rats fed with HFD (45% cal. from fat) stimulated with 25 mM ethanol	1, 3 or 5 days of stimulation	↑Supernatant levels of pro-inflammatory cytokines (TNFα, IL-6)	↑HSCs activation ↑oxidative stress
Vecchione G et al., 2016109	FaO hepatoma cells + ethanol + FFA	Rat hepatoma FaO cell line	FaO hepatoma cells + 100 mM etanol + 0.35 nM oleate/palmitate (2:1)	1 day of stimulation	—	↑Steatosis Altered lipid metabolism (↓ PPARγ and SIRT1) ↓Metallothionein antioxidant response

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CXCL, C-X-C motif chemokine ligand; ER, endoplasmic reticulum; FA, fatty acid; FFA, free fatty acids; GSH glutathione; GPX, glutathione peroxidase; GR, glutathione reductase; HFD, high-fat diet; IL, interleukin; LDH, lactate dehydrogenase; LPS, lipopolysaccharide; M1 iNOS, macrophages 1 inducible nitric oxide synthase; M2 Arg1, macrophages 2 arginase 1; OLETF, Otsuka Long Evans Tokushima Fatty; PCLS, precision-cut liver slices; PPAR, peroxisome proliferator-activated receptor; Sirt1, sirtuin 1; SOD, superoxide dismutase; STHD, steatohepatitis-inducing high-fat diet; TG, triglycerides; TNFa, tumour necrosis factor-α; WD, western diet.

Current efforts are focused on developing complex human-based preclinical models to underscore clinically relevant mechanisms and potential therapeutic targets. In this regard, patient specific in vitro models such as patient-derived and induced pluripotent stem cell-derived organoids and spheroids may be powerful systems for advancing our understanding of MetALD from a personalised point of view.

Implications for early detection and therapy

Moving towards early detection of MetALD

Alcohol and MRFs are major causes of liver diseases and are responsible for a large portion of the morbidity and mortality burden of liver diseases globally, and particularly in Europe and America¹. In the last years, we have experienced a change in the paradigm of the management of liver diseases, with international organisations and expert consensus recommending the development of programmes and initiatives for early detection and screening of liver diseases.¹¹² There is increasing evidence supporting the viability, acceptability and potential beneficial effects of screening programmes for liver diseases.¹¹³ Screening programmes may be untargeted, that is focusing on the entire population regardless of risk factors of liver disease, or targeted to specific populations at higher risk of liver disease. The latter approach is followed by the majority of international organisations and recommendations.¹⁰ As most recommendations for screening for liver disease derive from MASLD, obesity or diabetes guidelines, they mainly focus on screening populations with MRFs, and only a minority also recommend screening in people with high-risk alcohol consumption.^114,115 Therefore, there is a risk that individuals with high-risk alcohol consumption may be neglected by the current screening strategies and programmes.

In terms of MetALD, detection and screening should target the identification and monitoring of the risk factors of liver disease, which include both alcohol consumption and metabolic status (see below), and the early detection of liver disease per se. As liver fibrosis has been shown to be the main driver of liver disease progression both in MASLD and ALD^53,55 strategies and tools developed for early detection of liver disease have focused on the detection of early stages of fibrosis. In recent years there have been a considerable number of studies investigating the performance of non-invasive biomarkers for screening of liver fibrosis. Biomarkers may be classified into two groups: serum biomarkers, either patented and non-patented tests, and imaging-based techniques for assessment of liver fibrosis. In the latter group, the most widely explored and validated biomarker for detection of liver fibrosis due to alcohol and/or MRFs is liver stiffness assessed by TE (Fig. 2). This non-invasive tool has shown a very good performance for detection and diagnosis of liver fibrosis on the one hand,¹¹⁶ and also for prognostication and prediction of liver-related events on the other.¹¹⁷ It is likely that a good strategy for detection of liver fibrosis in patients at risk of liver disease would include a combination of non-patented, widely available and affordable serum tests, followed by a confirmatory, more accurate and more expensive test (such as TE or enhanced liver fibrosis test) in cases where there is a high suspicion of liver disease after the initial test.

Fig. 2.

Early detection of MetALD through targeted screening.

The first step for screening should be the assessment of risk factors of liver disease to identify the population at risk; the second step is the assessment of liver disease severity using serum biomarkers and transient elastography. Images from Servier Medical Art were used to create this figure. Servier Medical Art by Servier is licensed under a Creative Commons Atribution 3.0 Unported License (https://creativecommons.org/licenses/by/3.0/). APRI, AST-to-platelet ratio index; AST, aspartate aminotransferase; AUDIT, Alcohol Use Disorders Identification Test; ELF, enhanced liver fibrosis; EtG, ethyl glucuronide; FIB-4, fibrosis-4; GGT, gamma-glutamyltransferase; MCV, mean corpuscular volume; MetALD, metabolic dysfunction and alcohol-related liver disease; PEth, phosphatidylethanol; TLFB, timeline follow back.

Regarding the cost-effectiveness of screening programmes for liver disease, current evidence is based on studies using theoretical models of costs and potential effects of screening.¹¹⁸ Despite the limitations of the theoretical models, these studies have shown a clear beneficial effect of different screening strategies for liver diseases.

Finally, as ALD and MASLD are caused by modifiable lifestyle habits, a crucial question to answer is whether screening for liver disease may also affect lifestyle habits. Emerging data strongly suggest that screening programmes have beneficial effects as modifiers of alcohol consumption habits, increasing engagement to dishabituation programmes, reducing the amount of alcohol consumed and increasing alcohol abstinence, as well as improving the control of MRFs.^119,120 Prospective well-designed studies are warranted to confirm these potential beneficial effects.

Biomarkers of metabolic status and alcohol consumption

As patients with MetALD share characteristics and risk factors both with MASLD and ALD, biomarkers for detection of this condition should focus on the metabolic status and a precise assessment of past and current alcohol use. In terms of assessing the metabolic status, there are specific biomarkers that can be used to monitor metabolic-associated conditions, such as glycosylated haemoglobin and free insulin for diabetes, body mass index and waist circumference for obesity or cholesterol and triglyceride levels in the case of dyslipidemia.¹²¹ Regarding the history of alcohol consumption, there are questionnaires available in the primary care setting, such as the AUDIT (Alcohol Use Disorders Identification Test) or its concise version (AUDIT-c), that have been shown to be useful in identifying individuals with alcohol use disorder, and consequently at risk of ALD.¹²² However, assessment of alcohol consumption based solely on self-report may be inaccurate since a proportion of patients underreport alcohol consumption.^123,124 The use of other biomarkers such as urinary ethyl glucuronide (EtG) or phosphatidylethanol (PEth) to investigate the history of alcohol consumption is particularly helpful to overcome this issue.^125,126 This is relevant considering that a significant number of patients may be misclassified as having MASLD when they actually meet the criteria of MetALD or even ALD, which significantly changes their prognosis.¹²⁷ In fact, a recent study assessed the impact of unreported alcohol consumption using EtG determination in hair in a cohort of patients previously classified as having MASLD. The study found that 20% of these patients showed results of hair EtG suggestive of moderate alcohol consumption, meeting criteria for MetALD, and 10% had high-risk alcohol consumption, meeting criteria for ALD.¹²⁴ The development and validation of new cost-effective biomarkers that are easy to apply and reliable is anticipated in the coming years.

New models of care

The challenge of the increasing coexistence of MRFs and alcohol implies the need to combine the expertise of different specialists in order to offer the best treatment options for patients. The development of multidisciplinary clinics to identify, modify and control not only liver disease progression, but also the risk factors of liver disease, is becoming more and more recommendable and could greatly benefit patients with MetALD. These clinics should ideally include hepatologists, addiction specialists, endocrinologists, physiotherapists, specialised nurses and motivational therapists. Although the implementation of these multidisciplinary clinics requires a considerable amount of resources and effort, this model of care has shown beneficial effects for the management of patients with ALD¹²⁸ and MASLD,¹²⁹ as well as patients with other chronic conditions.¹³⁰ The implementation of new digital solutions through smartphone applications might also be useful for monitoring risk factors and suggesting interventions in patients with MetALD, as they require minimal active involvement from patients and may provide information in real-time. This approach has been shown to be effective in several health conditions, such as diabetes.¹³¹ These Apps have also shown promising results in patients with MASLD or ALD. A recent single arm study found a positive effect of a therapeutic intervention based on a digital App in patients with MASLD, improving histological NAFLD activity score in this cohort.¹³² Furthermore, a recent study including patients with ALD and alcohol use disorder (AUD) has shown data suggesting the possibility to estimate craving in this population using a smartphone App.¹³³ New digital solutions combining monitoring and personalised interventions may be very useful for patients with MetALD and are expected to be an important field of research in the near future.

Therapeutic interventions for MetALD

Like in any other health condition, the cornerstone in the treatment of MetALD should be the treatment of the aetiological factors of the disease. For patients with obesity, the first-line therapy is physical activity and dietary intervention, followed by bariatric surgery as second-line therapy when indicated. Patients with diabetes also need lifestyle modifications but often require medications to lower glucose. In patients with AUD, first-line therapy involves psychosocial interventions, followed by AUD medications to decrease alcohol consumption. Nevertheless, many patients with MetALD will require additional therapies at some point. To date, medications targeting pathophysiological mechanisms of disease progression in SLD, such as steatohepatitis or fibrosis, have mainly focused on MASLD. However, MASLD and ALD share many pathogenetic pathways; therefore, drugs investigated for MASLD could potentially be useful in ALD and MetALD as well. Depending on the mechanism of action of each drug, it is reasonable to expect lower, similar or even higher efficacy in patients with MetALD compared to those with MASLD. Fig. 3 shows the interventions available for the management of these patients, highlighting the most promising options.

Fig. 3.

Therapeutic interventions available for the management of patients with MetALD.

Lines encompass the spectrum where each intervention is likely to be useful. Green stars represent medications that have been investigated for MASLD but have also shown signs of reducing alcohol consumption in animal models and/or humans. Created in BioRender. Pose, E. (2024) https://BioRender.com/c13j905. ALD, alcohol-related liver disease; AUD, alcohol use disorder; FGF21, fibroblast growth factor 21; FXR, farnesoid X receptor; GLP1-RA, glucagon-like peptide-1 receptor agonist; MASLD, metabolic dysfunction-associated steatotic liver disease; MetALD, metabolic dysfunction and alcohol-related liver disease; PPAR, peroxisome proliferator-activated receptor; THR-β, thyroid hormone receptor-β.

Two groups of medications might be of particular interest for MetALD due to some evidence suggesting beneficial effects on AUD: i) the glucagon-like peptide-1 receptor agonists; and ii) the fibroblast growth factor 21 (FGF21) analogues. Semaglutide, a glucagon-like peptide-1 receptor agonist used for patients with type 2 diabetes and obesity that has been shown to reduce steatohepatitis in patients with MASLD,¹³⁴ has also been associated with a reduction in alcohol consumption in animal models¹³⁵ and in humans. A recent retrospective study in a large cohort of patients with obesity showed that semaglutide compared with other anti-obesity medications was associated with a 50-56% lower risk of both incidence and recurrence of AUD at 12 months. In addition, these results were replicated in a cohort of patients with type 2 diabetes.¹³⁶ The underlying mechanisms by which semaglutide may reduce alcohol consumption are likely a combination of central (i.e. modulation of the dopamine reward system)¹³⁷ and peripheral mechanisms (i.e. delayed gastric emptying and increased satiation). Of note, semaglutide should be used with caution in patients with normal body mass index, owing to its weight loss-inducing effect. Efruxifermin and pegozafermin, two FGF21 analogues showing positive results in MASH resolution and fibrosis regression^138,139 may also be of high interest for MetALD. FGF21 is a stress-induced hormone secreted from the liver and adipose tissue in response to high-fructose and -alcohol ingestion that has been shown to reduce alcohol consumption in animal models.¹⁴⁰ Data in humans are lacking but are expected in the coming years upon approval of FGF21 analogues for MASLD.

Drugs targeting HSC activation and fibrogenesis might have similar effects in patients with MASLD, ALD and MetALD, as liver fibrosis is the converging point of the disease-driving pathways of the three conditions. Similarly, medications targeting lipid and/or bile acid metabolism (i.e. obeticholic acid or lanifibranor) might also have similar effects not only on hepatic fat and inflammation, but also on fibrogenesis considering that lipid metabolism is a known trigger for HSC activation. This mechanism could explain the antifibrotic effects observed in phase II/III trials with these drugs.^141,142

In contrast to the medications discussed above, resmetirom might not be as effective in patients with MetALD. Resmetirom is a thyroid hormone receptor beta-selective agonist recently approved for the treatment of MASLD.¹⁴³ Thyroid hormones in the liver promote lipid oxidation, control hepatic insulin sensitivity and decrease hepatic gluconeogenesis via the thyroid hormone receptor, among other actions. Dysfunctions in this pathway have not been reported in ALD; therefore, the efficacy of resmetirom in MetALD may be lower, especially in patients who are ALD predominant.

Regarding medications tested in ALD, larsucosterol, an endogenous oxysterol that acts as an epigenetic regulator by inhibiting several DNA methyltransferases, showed promising efficacy results in a phase I/IIa trial in patients with alcohol-related hepatitis.¹⁴⁴ Results from a phase IIb trial are expected soon.

Surgical interventions also deserve a comment. In patients with MASLD and obesity, there is convincing evidence showing regression of liver disease and reduction in long-term mortality after bariatric surgery.^145,146 However, some studies have also shown an increase in AUD diagnoses after surgery and more frequently when performing bypass procedures compared to sleeve gastrectomy.¹⁴⁷ AUD in these patients has been associated with adverse long-term outcomes, especially in females¹⁴⁸ and in patients with binge drinking.¹⁴⁹ This is particularly concerning when considering patients with MetALD for bariatric surgery. Until more information becomes available, bariatric surgery seems a reasonable option, opting for sleeve gastrectomy when possible, in patients with MetALD who are MASLD predominant and have severe obesity, while a cautious case-by-case assessment is required for patients whose alcohol use is close to the threshold for ALD or in those with a binge drinking pattern before surgery. In all cases, we recommend the use of quantitative biomarkers of alcohol consumption and an assessment by addiction specialists before and after surgery.

Finally, given the complex pathophysiology of SLD, combination therapy is an appealing option that has shown preliminary signs of efficacy in MASLD,¹⁵⁰ and should be explored for patients with MetALD in the following years.

Future prospects

The new SLD nomenclature has had a great impact since its publication and has been the subject of several studies, conferences and symposia in hepatology and other specialties. It has also attracted great interest from the pharmaceutical industry.

Future studies evaluating the prognostic implications of the different categories of SLD and MetALD in particular are warranted. In this context, a precise assessment of alcohol consumption in this population should be established in prospective studies. First, as alcohol consumption plays a key role in the classification of SLD, some aspects will need further clarification: the role of a history of alcohol consumption in abstainers at the time of diagnosis, which is not considered in the current classification, certainly deserves special consideration. In addition, the specific windows to define current alcohol consumption need to be better defined based on high quality data. A dynamic, rather than static, classification of patients with MASLD, MetALD or ALD should be considered, as risk factors often vary over time. In this regard, the implementation of biomarkers of alcohol consumption such as PEth both in clinical practice and research, and a more precise definition of the thresholds for moderate and high-risk alcohol consumption that correlate with the MetALD and ALD categories will be very useful in classifying patients across the SLD spectrum. For this matter, we insistently recommend that all trials testing medications for SLD include the measurement of biomarkers for alcohol use, preferably PEth, at each study visit. This is crucial to generate robust data on the correlation between alcohol use biomarkers and SLD subtypes.

As discussed in this review, medications under development for MASLD target pathogenic pathways common to MetALD. Therefore, future clinical trials in SLD should consider including patients with MetALD. The inclusion of these patients would enhance enrolment, which is one of the most important challenges in the design of trials in MASLD. However, the negative impact that active alcohol consumption may have on the efficacy of the tested drugs will have to be accounted for in sample size calculations. Although stratification for alcohol intake seems reasonable in this setting, it may not be accurate as it is plausible that many patients with MetALD will stop consuming alcohol at inclusion because they may not even have an AUD. Other strategies to overcome the sample size limitation and differences in the study population may involve using adaptive trial designs, such as sample size re-estimation, adaptive randomisation or biomarker-adaptive designs. Moving from the classical analysis of a dichotomised outcome measure (i.e. improvement of at least one stage in fibrosis) to an ordinal analysis of outcomes (i.e. shift towards lower stage of fibrosis) may also help in reducing sample size requirements and avoiding the failure to find treatment differences upon reductions in power.

Regarding translational research, studies focusing on the pathogenesis and mechanisms of liver injury common to both MRFs and alcohol will move the field forward and promote the development of new drugs targeting both causes of liver disease. In the era of precision medicine, SLD and specifically MetALD might be the perfect setting for hepatology researchers and clinicians to develop personalised therapy.

Abbreviations

ALD, alcohol-related liver disease; AUD, alcohol use disorder; CAP, controlled attenuation parameter; FGF21, fibroblast growth factor 21; EtG, ethyl glucuronide; FXR, farnesoid X receptor; HCC, hepatocellular carcinoma; HSCs, hepatic stellate cells; MASLD, metabolic dysfunction-associated steatotic liver disease; MetALD, metabolic dysfunction and alcohol-related liver disease; MRI-PDFF, MRI-proton density fat fraction; MRFs, metabolic risk factors; NAFLD, non-alcoholic fatty liver disease; PEth, phosphatidylethanol; SLD, steatotic liver disease; TE, transient elastography; TGR5, Takeda G-protein coupled receptor 5.

Financial support

This work has been funded by Instituto de Salud Carlos III (ISCIII) through the project "PI022/00910" (PI: Elisa Pose) and co-funded by the European Union; this includes BioRender subscription of E.P.

Authors’ contributions

All authors contributed to writing original draft, as well as manuscript review and editing.

Conflict of interest

RB has received consulting fees from Novo Nordisk, GSK, Boehringer Ingelheim and Resolution therapeutics, and is on the speakers’ bureau for AbbVie and Gilead. All other authors report no conflicts.

Please refer to the accompanying ICMJE disclosure forms for further details.

Supplementary data

The following are the Supplementary data to this article: