Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) is defined by hepatic steatosis in the context of cardiometabolic risk factors, such as obesity, type 2 diabetes mellitus (T2DM), hypertension, and dyslipidemia, and may progress to metabolic dysfunction-associated steatohepatitis (MASH).1 MASLD is a heterogeneous disorder with a variable clinical trajectory, ranging from simple steatosis to MASH, advanced fibrosis, cirrhosis, and hepatocellular carcinoma.2,3 Driven in part by the global epidemics of obesity and T2DM, MASLD has become the most prevalent chronic liver disease worldwide, affecting approximately 1.3 billion individuals in 2023 and imposing a substantial global health burden.4,5 Moreover, the prevalence of MASLD is projected to continue rising over the coming decades.6

Pathogenetically, MASLD is a multifactorial disease driven by complex interactions among metabolic, inflammatory, genetic, and environmental factors.7,8 Importantly, concomitant alcohol consumption can aggravate metabolically driven liver injury by enhancing hepatic lipogenesis, impairing mitochondrial fatty acid oxidation, disrupting lipid handling, and promoting oxidative stress, thereby contributing to the development and progression of metabolic dysfunction and alcohol-associated liver disease (MetALD).9 The progression from simple liver steatosis to MASH results from excessive hepatic fatty acid uptake, increased de novo lipogenesis, and impaired mitochondrial β-oxidation.10 These alterations induce cellular stress and inflammation, which activate hepatic stellate cells (HSCs), promote fibrogenesis, and ultimately contribute to cirrhosis development.11 Crosstalk among hepatocytes, immune cells, and HSCs establishes a feed-forward cycle of immune activation, fibrogenesis, and tissue remodeling.12 Beyond local liver injury and inflammation, extrahepatic drivers, including gut-liver axis dysbiosis, systemic obesity, insulin resistance, and underlying cardiometabolic diseases, profoundly shape the hepatic microenvironment, accelerating MASLD progression.13

The multifactorial nature of MASLD complicates therapeutic development and underscores the need for personalized treatment strategies.14 Despite the recent accelerated approval of resmetirom and the rapid development of incretin- and glucagon-based therapies,15,16 MASH remains a major unmet medical need owing to its complex and heterogeneous pathogenesis. To date, numerous candidate therapies are currently under preclinical and clinical investigation, while precision delivery and cell-selective therapeutic technologies are emerging for MASLD/MASH.17

By adopting a liver cell-centered framework, this review highlights how cellular heterogeneity, intercellular crosstalk, and disease-stage dependency shape the efficacy and translational potential of emerging therapies for MASH, with particular emphasis on emerging experimental interventions and unresolved knowledge gaps across major hepatic cellular compartments (Figure 1 and Table 1). A more comprehensive understanding of these cellular networks may support the development of more effective and personalized therapeutic strategies for MASH.

Table 1 Summary of Key Experimental Therapeutics for Metabolic Dysfunction-Associated Steatohepatitis (MASH)

Figure 1 Key therapeutic network for metabolic dysfunction-associated steatohepatitis (MASH). Schematic overview of major experimental therapeutic strategies for MASH is organized according to their principal cellular and systemic targets. Hepatocyte-directed therapies aim to reduce steatosis, lipotoxicity, metabolic stress, and cell injury through modulation of thyroid hormone receptor-β, fibroblast growth factor signaling, peroxisome proliferator-activated receptors, farnesoid X receptor, de novo lipogenesis, mitochondrial function, senescence, and apoptosis/stress pathways. Non-parenchymal and immune cell-targeted approaches include therapies focusing on macrophages/Kupffer cells, lymphocytes, hepatic stellate cells, liver sinusoidal endothelial cells, and cholangiocytes, with the goal of attenuating inflammatory amplification, fibrogenesis, vascular dysfunction, and niche disruption. Furthermore, systemic therapies, including incretin- and glucagon-based agents, sodium-glucose cotransporter 2 inhibitors, lifestyle intervention, and bariatric approaches, act across extrahepatic and intrahepatic compartments to improve metabolic homeostasis and indirectly reduce hepatic inflammation and fibrosis. Additionally, precision delivery and cell-selective technologies, including GalNAc-conjugated nucleic acid therapeutics, lipid nanoparticles, vitamin A-coupled liposomes, and engineered cell therapies, represent emerging therapeutic strategies to improve cellular specificity and efficacy. Arrows may only indicate major cell specificity, as many agents exert multicellular or systemic effects. Schematic illustrations were created using BioRender.

Hepatocyte-Targeted Therapies: Modulating Steatosis, Lipotoxicity and Cell Death

Hepatocytes perform the majority of hepatic functions, including the metabolism of glucose, lipids, proteins, and bile acids, as well as detoxification.76,77 In the healthy liver, lipid homeostasis is maintained through a balance among fatty acid uptake, de novo lipogenesis, fatty acid oxidation, triglyceride storage, and very-low-density lipoprotein (VLDL) export.7,78 In MASLD, excessive intrahepatic triglyceride accumulation triggers hepatocyte death, thereby driving disease progression to MASH.8 Accordingly, multiple hepatocyte-targeted therapeutic strategies are currently being investigated.

Thyroid Hormone Receptor-β (THR-β) Agonists

THR-β agonists represent one of the most advanced hepatocyte-directed therapeutic classes for MASH.79 By selectively activating hepatic THR-β signaling, these agents enhance mitochondrial function and β-oxidation, reduce hepatic steatosis and lipotoxic stress, and may thereby indirectly attenuate HSC activation and fibrogenesis.80 In the phase III MAESTRO-NASH trial, 52 weeks of resmetirom treatment significantly improved both major histological endpoints. MASH resolution, defined as a ballooning score of 0, a lobular inflammation score of 0 or 1 and a ≥2-point reduction in NAFLD Activity Score (NAS), without worsening of fibrosis, was achieved in 25.9% and 29.9% of patients receiving 80 mg and 100 mg resmetirom, respectively, compared with 9.7% receiving placebo. Furthermore, ≥1-stage fibrosis improvement without worsening of MASH occurred in 24.2% and 25.9% of the respective resmetirom groups versus 14.2% with placebo.18 Resmetirom is the first THR-β agonist that has received accelerated approval by the US Food and Drug Administration (FDA) and conditional approval by the European Medicines Agency (EMA) for treating MASH with F2 or F3 fibrosis,81 while its efficacy in MASH-related compensated cirrhosis is currently under investigation in an ongoing randomized phase III trial (ClinicalTrials.gov - NCT05500222).

Beyond resmetirom, other selective THR-β agonists, such as VK2809 and TERN-501, have shown promising results in Phase 2 clinical trials, demonstrating significant reductions in liver fat content and improvements in fibrosis in patients with MASH.82 VK2809 is a liver-targeted THR-β agonist prodrug that undergoes CYP3A4-mediated activation in the liver, resulting in high hepatic selectivity. In the phase IIb VOYAGE trial, VK2809 significantly reduced liver fat content and improved histological outcomes, including MASH resolution and fibrosis improvement, with a favorable safety profile.19 TERN-501 demonstrated significant dose-dependent reductions in liver fat content and a favorable safety profile in the phase IIa DUET study, suggesting its therapeutic potential for MASH.20 Additionally, several next-generation THR-β agonists, including ASC41, IS25 and TG68, are also currently under development.83

Although THR-β agonists have demonstrated efficacy in MASH, several important issues remain to be addressed. First, clinical trial data indicate that resmetirom is a generally well tolerated; nevertheless, appropriate safety monitoring remains necessary during treatment.18 Particular attention should be paid to hepatic function, thyroid status, and cardiovascular and skeletal safety. In patients receiving concomitant statin therapy, dose adjustments should follow prescribing recommendations, with subsequent reassessment of lipid profiles.84,85 Second, the optimal duration of THR-β agonist therapy and the durability of therapeutic benefits after treatment discontinuation remain to be determined.86,87 Third, a considerable proportion of patients fail to achieve meaningful histological improvement, highlighting substantial inter-individual variability in therapeutic response. Therefore, predictors of efficacy, including potential genetic factors such as THRB, SLCO1B1, and PNPLA3 variants, require further investigation.18 Finally, the cost and accessibility of THR-β agonists may limit their widespread implementation, particularly in resource-limited healthcare settings.20 Further studies are needed to establish long-term efficacy and safety, optimize treatment duration, and identify patients most likely to benefit from THR-β agonist therapy.

Fibroblast Growth Factor (FGF)-Based Therapies

FGF-based therapies represent another strategy with prominent hepatic metabolic effects in MASH, with FGF21 analogues and FGF19 mimetics showing beneficial effects on steatosis, metabolic dysfunction, and fibrosis.88 Several FGF21 analogues, including efruxifermin, pegozafermin, and efimosfermin, have demonstrated promising efficacy in patients with MASH, reducing liver fat content, improving insulin sensitivity and lipid profiles, and increasing rates of MASH resolution and fibrosis improvement compared with placebo.89 These agents are generally well tolerated, with gastrointestinal symptoms being the most common adverse events.23,24 Although efruxifermin did not significantly improve the primary fibrosis endpoint at week 36 in the phase IIb SYMMETRY trial, longer-term treatment suggested increasing antifibrotic efficacy in patients with MASH-related compensated cirrhosis. At week 36, ≥1-stage fibrosis improvement without worsening of MASH occurred in 18% and 19% of patients receiving 28 mg and 50 mg efruxifermin, respectively, compared with 13% receiving placebo. By week 96, the corresponding response rates increased to 21% and 29%, respectively, versus 11% with placebo, suggesting that prolonged FGF21 agonism may be required to achieve meaningful fibrosis regression in advanced MASH.90 Similarly, longer-term follow-up from the HARMONY study in patients with F2/F3 fibrosis suggested durable fibrosis improvement.91 Mechanistically, preclinical data indicate that efruxifermin may exert direct antifibrotic effects independent of changes in steatosis or inflammation, whereas its clinical benefits are generally interpreted as consequences of broader metabolic improvement.92 Looking ahead, several ongoing phase III clinical trials are evaluating the efficacy and safety of efruxifermin, pegozafermin and efimosfermin (ClinicalTrials.gov: NCT06215716, NCT06419374, NCT06318169, and NCT07221188).

Aldafermin, the most extensively studied FGF19 mimetic, demonstrated a dose-dependent reduction in liver fat content and showed favorable trends toward MASH resolution and fibrosis improvement in phase II studies.93,94 In patients with compensated MASH cirrhosis, aldafermin also reduced enhanced liver fibrosis (ELF) scores, implying potential antifibrotic activity in advanced disease.26 However, aldafermin treatment was associated with dose-dependent increases in low-density lipoprotein (LDL) cholesterol, which could be effectively controlled with concomitant statin therapy.95

Peroxisome Proliferator-Activated Receptor (PPAR) Agonists

PPARs are nuclear receptors that regulate lipid metabolism, glucose homeostasis, and inflammation. The three major isoforms, PPARα, PPARγ, and PPARδ, have been targeted by agonists that improve hepatic steatosis and metabolic dysfunction by enhancing fatty acid oxidation, reducing lipotoxicity, and modulating inflammatory signaling, thereby representing attractive therapeutic targets for MASH.96

Pioglitazone, a PPARγ agonist, has been demonstrated to improve steatosis and inflammation with potential benefits for fibrosis. However, concerns regarding weight gain, fluid retention, and cardiovascular safety have limited its broader clinical use.97 Accordingly, novel pioglitazone derivatives are being developed to improve safety while preserving therapeutic efficacy.33

To enhance therapeutic efficacy, dual- and pan-PPAR agonists have been developed to simultaneously modulate multiple metabolic and inflammatory pathways involved in MASH. The PPARα/δ agonist elafibranor showed encouraging results in early-phase studies but failed to meet primary endpoint in a phase III trial27 (NCT02704403). In contrast, the dual PPARα/γ agonist saroglitazar has demonstrated encouraging metabolic and histological benefits in early clinical studies.28,33 Among these agents, the pan-PPAR agonist lanifibranor has shown the most promising clinical profile, significantly improving both MASH resolution and fibrosis in a phase II trial and advancing to phase III evaluation with phase III efficacy data expected by the end of 2026.30

Nevertheless, several challenges remain for PPAR-based therapies in MASH, including limited long-term efficacy and safety data, uncertainties regarding cardiovascular safety across different agents, and adverse effects associated with PPARγ activation, such as weight gain and edema, underscoring the need for careful patient selection and monitoring.30,98

Farnesoid X Receptor (FXR) Agonists

Farnesoid X receptor is a nuclear receptor that plays a central role in regulating bile acid, lipid, and glucose metabolism. FXR activation can reduce hepatic steatosis, suppress inflammation, attenuate hepatocellular injury and death, and inhibit fibrogenesis. Given these pleiotropic effects, FXR agonists have been developed and evaluated as potential therapies for MASH.99,100

Obeticholic acid (OCA), a semisynthetic bile acid derivative and FXR agonist, is the most extensively studied agent in this class.101 In the phase II FLINT trial, 72 weeks of OCA treatment resulted in ≥1-stage fibrosis improvement in 35% of patients compared with 19% receiving placebo.102 These findings were subsequently supported by the phase III REGENERATE trial, in which ≥1-stage fibrosis improvement without worsening of MASH after 18 months occurred in 22.4% of patients receiving OCA 25 mg compared with 9.6% receiving placebo.34 However, OCA did not significantly improve the co-primary endpoint of MASH resolution, and the phase III REVERSE trial in patients with compensated MASH cirrhosis failed to meet its primary endpoint.103

Although OCA has demonstrated antifibrotic activity, its overall clinical benefit has been modest, and accompanied by important tolerability and safety concerns, including pruritus, cholelithiasis, potential hepatotoxicity, and unfavorable lipid-profile changes characterized by increased LDL cholesterol and reduced HDL cholesterol. Whether these limitations are specific to OCA or reflect broader class effects of FXR agonism remains unclear.103–105 Consequently, the FDA determined that the overall benefit-risk profile of OCA did not support approval for MASH.99

In addition to OCA, several non-steroidal FXR agonists, including cilofexor,35 tropifexor,36 vonafexor,37 and EDP-305,38 have shown promising metabolic and hepatic effects in phase II trials. However, their effects on fibrosis have generally been modest, while pruritus remains a frequent tolerability concern across several FXR agonists.101 In particular, the precise mechanism of pruritus remains unclear, but OCA-induced alterations in bile-acid homeostasis may modify exposure to pruritogenic bile-acid species and activate sensory pathways implicated in cholestatic itch, including MRGPRX4 signaling.106 Future development of FXR agonists may prioritize greater target selectivity, optimized pharmacokinetic profiles, and improved tolerability to enhance efficacy while minimizing adverse effects such as pruritus; alternative approaches to modulating FXR signaling and rational combination therapies may further improve therapeutic outcomes in MASH.

Lipogenesis-Targeted Therapies Acetyl-CoA Carboxylase (ACC) Inhibitors

De novo lipogenesis (DNL) is a major contributor to hepatic triglyceride accumulation in MASH.107,108 Excessive fatty acid synthesis promotes lipotoxicity, hepatocellular injury, inflammation, and subsequent fibrogenesis.96 ACC catalyzes the conversion of acetyl-CoA to malonyl-CoA and plays a central role in regulating both fatty acid synthesis and β-oxidation, making it an attractive therapeutic target for MASH.109–111 Preclinical studies have demonstrated that ACC inhibitors reduce liver inflammation, improve liver function and insulin resistance, and exert antifibrotic effects in animal models.112–114 In clinical studies, dual ACC1/2 inhibitors such as firsocostat115 and clesacostat40 have been shown to reduce liver fat content and DNL activity, whereas convincing evidence of fibrosis improvement remains limited.

Despite encouraging effects on hepatic steatosis, ACC inhibitors face several important limitations. First, safety concerns remain, particularly hypertriglyceridemia, which may increase cardiometabolic risk and the risk for acute pancreatitis, as well as reduced platelet counts observed in some studies. Second, the role of DNL in MASLD is not fully understood, and it remains unclear whether DNL dysregulation is a universal feature of the disease or is restricted to specific patient subsets. Third, the relative contributions of ACC1 and ACC2 to disease pathogenesis and treatment response remain poorly defined, while most available evidence is derived from dual ACC1/2 inhibition. Interestingly, preclinical studies also suggest that ACC inhibition can directly suppress stellate cell activation, indicating potential antifibrotic effects beyond hepatocyte metabolic modulation.116 Future studies in larger and more diverse patient populations are needed to define optimal patient subsets and determine whether isoform-selective inhibitors and rational combination therapies can improve efficacy while minimizing adverse effects.39,108,111

ATP-Citrate Lyase (ACLY) Inhibition

ATP-citrate lyase (ACLY), generating cytosolic acetyl-CoA from citrate and thereby supplies substrate for de novo lipogenesis and cholesterol synthesis, represents another emerging metabolic target in MASLD/MASH. Bempedoic acid is an orally administered ACLY inhibitor that undergoes preferential activation in the liver and is clinically used as an LDL-cholesterol-lowering therapy, including in patients with statin intolerance.117 Mechanistically, bempedoic acid reduced fibrogenic transcriptional programs and directly influenced hepatic stellate cell activation, indicating that its therapeutic effects may extend beyond suppression of hepatocyte lipogenesis.41 More recent work has confirmed potent suppression of hepatic de novo lipogenesis and steatosis by bempedoic acid, although these effects may not be explained exclusively by hepatic ACLY inhibition.42 Furthermore, combining ACLY inhibition with GLP-1 receptor agonism produced additive improvements in steatohepatitis and hepatic fibrosis in experimental models, suggesting potential utility in combination strategies.118 Nevertheless, clinical evidence supporting bempedoic acid as a direct treatment for MASLD/MASH remains limited, and its histological efficacy in patients has yet to be established.

Fatty Acid Synthase (FASN) Inhibitors

FASN, a key regulator of DNL, represents a promising therapeutic target for MASLD. Denifanstat (TVB-2640) is an oral FASN inhibitor that suppresses DNL and reduces the generation of lipotoxic fatty lipid species implicated in inflammation, hepatocellular injury, and fibrogenesis in MASH.119 In the phase II FASCINATE-1 study, denifanstat reduced liver fat content in a dose-dependent manner, improved metabolic, inflammatory, and fibrotic biomarkers, with a generally favorable safety profile.120 These findings were further supported by the phase IIb FASCINATE-2 trial in patients with biopsy-confirmed MASH and F2–F3 fibrosis. Treatment with denifanstat resulted in a ≥2-point reduction in NAS without worsening fibrosis in 38% of patients, compared with 16% in the placebo group. In addition, MASH resolution without worsening fibrosis was achieved in 26% versus 11% of patients, respectively. Denifanstat was generally well tolerated, with alopecia and dry eye symptoms reported as the most common treatment-related adverse events.43

Despite these encouraging results, the subsequent phase III development program was discontinued for reasons that have not been publicly disclosed; consequently, the long-term efficacy, safety, and clinical utility of FASN inhibition in MASH remains to be established.121 Beyond denifanstat, several novel FASN inhibitors are under development. For example, 84-B10 reduced hepatic lipid accumulation and improved metabolic abnormalities in high-fat diet (HFD)-fed mice by promoting FASN degradation, further supporting FASN as a therapeutic target in MASLD.44

Diacylglycerol O-Acyltransferase-2 (DGAT2) Inhibitor

DGAT2 catalyzes the esterification of fatty acids with diacylglycerol and is a rate-limiting enzyme in triglyceride synthesis. DGAT2 inhibition reduces fatty acid synthesis and lowers triglyceride (TG) accumulation and secretion from liver steatosis.122

Ervogastat (PF-06865571) is a first-in-class oral DGAT2 inhibitor that has shown dose-dependent reductions in liver fat content and serum triglyceride levels.123,124 In the phase II MIRNA trial involving patients with biopsy-confirmed MASH and F2–F3 fibrosis, ervogastat, either as monotherapy or in combination with the ACC inhibitor clesacostat, produced favorable histological effects after 48 weeks of treatment. Although combination therapy appeared to confer greater hepatic benefit, it was associated with potentially unfavorable changes in lipid and apolipoprotein profiles. Overall, these findings support further evaluation of ervogastat as a therapeutic candidate for MASH with fibrosis.45 ION224, a liver-targeted DGAT2 antisense oligonucleotide, significantly improved a ≥2-point reduction in NAS and fibrosis in a phase II trial while maintaining a favorable safety profile, without hypertriglyceridemia or adverse lipid-profile changes. However, interpretation of these findings is limited by the relatively small sample size, one-year follow-up, and lack of evaluation in patients with cirrhosis.46 Further studies in larger and more diverse patient populations are needed to confirm the long-term efficacy and safety of ION224.

Mitochondrial and Endoplasmic Reticulum (ER) Stress-Targeted Therapies

Mitochondrial dysfunction and ER stress represent crucial metabolic components in line with hepatocyte lipotoxicity, therefore emerging as additional therapeutic targets in MASH.7 Persistent lipid (often together with alcohol in adults) overload can impair mitochondrial oxidative capacity and quality control, increase reactive oxygen species production, and promote mitochondrial danger-signal release, while simultaneously triggering ER stress and maladaptive unfolded protein response (UPR) signaling.125 Together, these processes amplify hepatocyte injury, inflammation, and fibrogenesis. Mitochondria-directed strategies include controlled mitochondrial uncoupling with HU6 and restoration of NAD+ metabolism through ACMSD inhibition, as discussed above, while modulation of mitochondrial biogenesis, mitophagy, and redox homeostasis remains predominantly at the preclinical stage.48,126 ER stress-directed approaches are less clinically mature but include chemical chaperones such as tauroursodeoxycholic acid (TUDCA), which attenuated PERK-CHOP-associated ER stress and MASH progression in experimental models, as well as emerging inhibition of the ER stress sensor IRE1α, which has been shown to reduce pyroptotic signaling and hepatic fibrogenesis in preclinical studies.127–129 Even though mitochondria-targeted therapies have achieved clinical success, THR-β agonists and FGF21-based agents may also indirectly improve mitochondrial and ER homeostasis by reducing lipid overload and restoring hepatocyte metabolic competence.77,130 Nevertheless, it requires more evidence and understanding on whether direct targeting of mitochondrial or ER stress pathways provides added clinical benefit beyond upstream metabolic modification.

Others

Beyond ACC, FASN, and DGAT2 inhibitors, several other metabolic therapies have exerted favorable effects on MASH. HU6 is a liver-activated controlled metabolic accelerator that generates the mitochondrial uncoupler 2,4-dinitrophenol, thereby increasing substrate utilization and promoting oxidation of fat and other carbon sources rather than their accumulation.47 An ongoing clinical trial is evaluating the safety, pharmacokinetics, and efficacy of HU6 in patients with MASH (ClinicalTrials.gov: NCT07491458). Another novel therapy, α-amino-β-carboxymuconate-ε-semialdehyde decarboxylase (ACMSD) inhibition to restore hepatic NAD⁺ metabolism, has shown beneficial effects on steatosis, inflammation, and mitochondrial function in cellular and animal models.48 Several emerging hepatocyte-directed approaches have shown promising results in recent preclinical studies. Among them, RNA-based therapies, such as hepatocyte-directed GalNAc-siTAZ, have demonstrated antifibrotic efficacy in experimental MASH models.70 Targeting senescent hepatocytes has also emerged as a potential therapeutic strategy, with senolytic interventions improving liver pathology and systemic metabolic dysfunction in preclinical models.131 Together, these findings expand the therapeutic landscape beyond conventional metabolic targets and may create new opportunities for precision treatment of MASH.

Beyond metabolic regulation, hepatocyte stress responses have also emerged as potential therapeutic targets. Agents such as the apoptosis signal-regulating kinase 1 (ASK1) inhibitor selonsertib and the pan-caspase inhibitor emricasan were developed to reduce hepatocellular injury and apoptosis. Despite encouraging preclinical data, clinical outcomes have been largely disappointing; trials of selonsertib did not demonstrate significant antifibrotic efficacy in patients with advanced MASH, leading to discontinuation of its further development.49 Similarly, the pan-caspase inhibitor emricasan did not improve histological, biochemical, or clinical outcomes across multiple clinical trials in patients with MASH-related fibrosis or cirrhosis.50,51,132 These disappointing results underscore the complexity of hepatocyte stress pathways and suggest that alternative hepatocyte-directed therapeutic strategies may be required.

In addition, targeting hepatocyte senescence has also emerged as a potential therapeutic strategy. Senescent hepatocytes accumulate with MASLD progression and exhibit a senescence-associated secretory phenotype (SASP) that can propagate metabolic dysfunction, inflammation, and fibrogenic signaling to neighboring hepatic cells.133 Recent cell-resolved analyses identified a senescent hepatocyte gene signature associated with disease severity and adverse clinical outcomes, while selective elimination of senescent hepatocytes ameliorated hepatic inflammation, fibrosis, and systemic metabolic dysfunction in experimental MASLD models.131

Immune Cell-Targeted Therapies: Dampening Inflammatory Amplification

Chronic inflammation represents a central driver of MASH progression and a major mechanistic bridge linking hepatocyte lipotoxicity to fibrogenesis. Injured hepatocytes release damage-associated molecular patterns (DAMPs), mitochondrial DNA, oxidized lipids, and inflammatory cytokines, which stimulate resident KCs and recruit monocyte-derived macrophages (MoMFs) into the liver. These innate immune responses subsequently amplify inflammatory signaling through cytokine production, inflammasome activation, and extensive crosstalk with HSCs, thereby promoting extracellular matrix deposition and fibrosis progression. Adaptive immune populations further contribute to chronic inflammatory remodeling and tissue injury persistence. Consequently, immune-targeted therapeutic strategies have emerged as attractive experimental approaches for MASH treatment.12 However, despite biological rationale, most immune-directed therapies have shown limited clinical efficacy in patients, highlighting substantial knowledge gaps regarding immune-cell heterogeneity, temporal disease dynamics, and inflammatory redundancy in MASH.

Macrophage/Kupffer Cell-Targeted Therapeutics and Unresolved Challenges

Macrophages stand at a central position in MASH pathogenesis owing to their roles in inflammatory modulation, lipid sensing, fibrogenic signaling, and tissue remodeling. Hepatic macrophage populations comprise both resident KCs and infiltrating MoMFs, which exhibit remarkable phenotypic plasticity during disease progression.134–136 Recent single-cell transcriptomic analyses have further identified distinct macrophage subsets, including TREM2+ macrophages enriched within fibrotic and steatotic niches.137,138 Nevertheless, the dynamic and diverse functions of these macrophage populations are not fully understood, representing a major obstacle for selective immunotherapeutic development.

One major therapeutic strategy aims to suppress inflammatory monocyte recruitment into the injured liver. The CCR2/CCR5 antagonist cenicriviroc was developed to inhibit chemokine-mediated recruitment and activation of inflammatory macrophages. Preclinical studies demonstrated attenuation of hepatic inflammation and fibrosis following CCR2/CCR5 blockade,139,140 while early clinical studies suggested antifibrotic activity in patients with MASH.141 However, the subsequent phase III trial failed to demonstrate clinical efficacy.52 These findings point out a major translational challenge in MASH immunotherapy: inflammatory signaling networks are highly redundant and dynamically regulated; therefore, such single-pathway blockades may be insufficient upon chronic fibrogenic establishment. Moreover, the optimal timing for macrophage-targeted intervention remains unclear, as inflammatory macrophages may exert distinct pathogenic or restorative functions depending on disease stage.134

Inflammasome signaling, a fundamental inflammatory driver in macrophages, has emerged as another important therapeutic target linking metabolic stress to innate immune activation.142 The NOD-like receptor pyrin domain-containing protein 3 (NLRP3) inflammasome promotes maturation of interleukin-1β (IL-1β) and IL-18 in response to lipotoxicity, mitochondrial reactive oxygen species (ROS), cholesterol crystal accumulation, and hepatocyte damage.143 Furthermore, inflammasome activation may induce pyroptotic cell death through Gasdermin D signaling, thereby amplifying hepatic inflammation.144 Till now, several NLRP3 inhibitors, including MCC950, DFV890, and inzomelid, have entered preclinical or early clinical development for inflammatory diseases;145–147 however, they have not yet focused on MASH. Furthermore, anakinra, an IL1 receptor antagonist (IL1RA) approved for inflammatory diseases, may improve outcomes of patients with severe alcohol-associated hepatitis,54 whereas it has been known to reduce insulin resistance in an early clinical trial.53 However, long-term suppression of inflammasome signaling may potentially interfere with host defense and tissue repair pathways.143 Hence, it is unclear whether inflammasome inhibition alone can sufficiently interrupt the inflammatory redundancy and fibrotic progression in advanced MASH.142

Free fatty acid receptor (FFAR) signaling is classically linked to anti-inflammatory effects, and hepatic FFAR4 expression has been reported to be primarily found in Kupffer cells. This makes liver macrophages a plausible major cellular target for its anti-inflammatory activity.148 In the phase IIb ICONA trial, icosabutate, a FFAR1/FFAR4 agonist, did not significantly improve MASH resolution without fibrosis worsening, but was associated with higher rates of fibrosis improvement, reduced biomarkers of liver injury, inflammation and glycaemic dysfunction, and a favorable safety profile. These findings support further evaluation of FFAR1/FFAR4 agonism as a potential antifibrotic and metabolic therapeutic strategy for MASH, particularly in patients with advanced fibrosis and type 2 diabetes.55 Future studies should clarify whether FFAR1/FFAR4 agonism is best positioned as monotherapy or as part of combination regimens targeting complementary metabolic, inflammatory and fibrotic pathways.

Gut-liver immune signaling has emerged as a considerable therapeutic interest. Increased intestinal permeability and microbiota dysbiosis enhance portal delivery of lipopolysaccharide (LPS) and other pathogen-associated molecular patterns (PAMPs) that activate Toll-like receptor 4 (TLR4) signaling in KCs.76 TLR4 activation subsequently promotes NF-κB-dependent inflammatory cytokine production and profibrotic signaling.149 TLR4 antagonists including eritoran149 and JKB-121 (ClinicalTrials.gov - NCT02442687) have been explored as experimental immunomodulatory therapies. Nevertheless, clinical translation has remained limited, likely due to involvement of multiple overlapping pathways in gut-liver immune interactions. On the other hand, it lights up the fact that microbiota-associated immune activation, indirectly responding to liver metabolic dysfunction, may not primarily drive fibrosis progression.

Cellular senescence may represent an additional layer of macrophage heterogeneity during chronic liver injury. Senescence- and aging-associated alterations in hepatic macrophages can impair homeostatic and reparative functions while favoring persistent inflammatory signaling.150 However, the specific contribution of senescent macrophage populations to MASH progression and their suitability for selective therapeutic targeting remain incompletely defined.

More recently, engineered immune-cell therapies have emerged as innovative experimental approaches for fibrosis modulation. Autologous macrophage infusion strategies aim to restore pro-resolving macrophage phenotypes capable of enhancing tissue repair and fibrosis regression.151 Such elaborate cell-based therapies may find a place in advanced stages such as liver cirrhosis.152 In parallel, chimeric antigen receptor (CAR) macrophages have demonstrated preclinical potential for targeted clearance of fibrogenic cells and extracellular matrix components.73,153 Although highly promising, substantial challenges remain regarding delivery efficiency, persistence, safety, manufacturing complexity, and functional stability within chronically inflamed hepatic microenvironments.

Adaptive Immune Modulation and Emerging Knowledge Gaps

In addition to innate immune activation, adaptive immune dysregulation increasingly appears to contribute to chronic inflammatory remodeling in MASH. Hepatic accumulation of activated CD8+ T cells, natural killer T (NKT) cells, mucosal-associated invariant T (MAIT) cells, and other lymphocyte populations has been associated with hepatocyte injury, inflammatory cytokine production, and fibrosis progression. Conversely, impaired regulatory T-cell (Treg) activity may further exacerbate chronic hepatic inflammation and tissue remodeling.154,155 Therefore, growing interest is prompted in immunomodulatory strategies targeting adaptive immune pathways. About ten years ago, foralumab, a monoclonal anti-CD3 antibody, was investigated in patients with MASH and T2DM but was suspended and withdrawn (ClinicalTrials.gov - NCT03291249). CAR T cells targeting senescent HSCs and liver fibrogenesis have demonstrated encouraging antifibrotic effects in preclinical models.74,75 In parallel, Treg-directed approaches including low-dose interleukin-2 administration, adoptive Treg transfer, and emerging CAR-Treg strategies have demonstrated immunomodulatory potential in inflammatory disease models.156,157 Additional experimental approaches targeting leukocyte recruitment and endothelial immune signaling, such as PXS-4728A (a VAP-1/SSAO inhibitor), have also been investigated for their anti-inflammatory and antifibrotic effects in the lung.158 Simultaneously, amine oxidase copper-containing 3 (AOC3) inhibitor, BI 1467335, attenuates leukocyte recruitment and provides dose-dependent reduction in the levels of liver injury biomarkers, alanine aminotransferase (ALT) and CK-18 in Phase IIa clinical trial.57 In addition, emerging preclinical strategies further include modulation of NKT- and MAIT-cell activity, chemokine signaling pathways such as the CXCR6/CXCL16 axis, and cytokine-directed therapies targeting IL-17, IL-1β, or JAK/STAT signaling pathways.159,160

Nonetheless, the translational potential of adaptive immune-targeted therapies in MASH remains uncertain owing to the multifaceted functions of individual lymphocyte subsets, which are highly context-, stage- and microenvironment-dependent. Furthermore, many adaptive immune abnormalities may arise during persistent metabolic injury, potentially compromising the efficacy of single-target immunomodulatory interventions. Future progress will likely require improved characterization of immune-cell heterogeneity with spatial location and time courses, alongside integrating metabolic, antifibrotic, and immune-targeted therapies.

Hepatic Stellate Cell (HSC)-Targeted Antifibrotic Therapies: Direct Approach with Formidable Hurdles

HSCs represent central drivers for hepatic fibrogenesis in MASH through excessive extracellular matrix (ECM) production and tissue remodeling. Following chronic hepatocyte injury and inflammatory stimulation, quiescent HSCs undergo activation into proliferative and contractile myofibroblast-like cells characterized by enhanced collagen deposition, matrix crosslinking, and profibrotic cytokine secretion.161 Hence, direct targeting of HSC activation and fibrogenic signaling has emerged as a promising therapeutic strategy in MASH.

Integrin-mediated activation of transforming growth factor-β (TGF-β) signaling represents one major antifibrotic target.162 Integrins such as αvβ1 and αvβ6 promote activation of latent TGF-β within fibrotic niches and contribute to HSC activation and matrix remodeling.163 Meanwhile, bexotegrast (PLN-74809), a dual αvβ1/αvβ6 integrin inhibitor, has therefore been investigated as a potential therapy for liver fibrosis in primary sclerosing cholangitis (PSC).164 However, systemic blockade of TGF-β related mechanisms may produce substantial off-target effects owing to its pleiotropic functions in immune regulation, tissue repair, and tumor suppression. The narrow therapeutic window frequently concerns the efficacy of antifibrotic therapies targeting TGF-β mechanisms. Furthermore, galectin-3, a β-galactoside-binding lectin involved in macrophage-HSC crosstalk, inflammatory amplification, and fibrogenesis, has attracted considerable therapeutic interest.165 Accordingly, belapectin (GR-MD-02), a galectin-3 inhibitor, demonstrated encouraging antifibrotic activity in preclinical studies and selected patient populations, although broader clinical efficacy on cirrhosis and portal hypertension remains limited.58 These findings suggest that interruption of isolated profibrotic signaling pathways may be insufficient once advanced fibrotic remodeling becomes established. Additionally, Hedgehog signaling contributes to HSC activation, tissue remodeling, and fibrotic niche formation in chronic liver injury, prompting investigation of hedgehog inhibitors such as vismodegib,166 sonidegib167 and LDE225 (ClinicalTrials.gov - NCT02151864).

Alternative strategies have focused on extracellular matrix stabilization and collagen processing. Lysyl oxidase-like 2 (LOXL2), an enzyme involved in collagen crosslinking and matrix stiffening, was targeted by the monoclonal antibody simtuzumab.168 However, despite promising experimental data, simtuzumab failed to demonstrate meaningful antifibrotic efficacy in clinical trials,59 highlighting the critical challenge that improvements in preclinical fibrosis models often fail to translate into clinical benefit in advanced human disease. In addition, HSP47-targeted siRNA therapy (BMS-986263) represents a more direct strategy to suppress collagen biosynthesis through inhibition of a collagen-specific molecular chaperone.169

Beyond direct antifibrotic agents, several metabolic therapies may indirectly modify HSC activation and fibrosis progression through regulation of hepatocyte stress, inflammatory signaling, and fibrogenic pathways.162,170 FXR agonists, including obeticholic acid, cilofexor, and tropifexor, have demonstrated antifibrotic potential through modulation of bile acid metabolism, suppression of inflammatory signaling, and inhibition of HSC activation.100 Similarly, FGF-based therapies, particularly FGF19 and FGF21 analogues (ie, aldafermin, efruxifermin, and pegozafermin) may attenuate fibrogenesis by improving metabolic homeostasis and reducing lipotoxic and inflammatory stimuli that drive HSC activation.171,172 PPAR agonists (ie, lanifibranor, pioglitazone, and saroglitazar) have also shown potential antifibrotic effects, consequentially through regulation of lipid metabolism, inflammation, and HSC trans-differentiation.30,173–176 Nevertheless, whether the fibrosis improvements under clinical observation primarily result from direct HSC modulation or secondary metabolic correction remains incompletely understood.

As is known, cellular senescence facilitates further complexity to HSC biology in MASH. Senescent HSCs are expanded in both human and experimental MASH and arise predominantly from activated HSC populations. Although senescence-associated proliferative arrest may limit further expansion of fibrogenic HSCs, persistent senescent HSCs can acquire a pro-inflammatory SASP and contribute to chronic inflammatory and fibrotic niche remodeling.177 This context-dependent behavior suggests that therapeutic induction or elimination of HSC senescence requires careful consideration of disease stage and cellular phenotype.

Collectively, the limited success of HSC-targeted antifibrotic therapies highlights several unresolved challenges in MASH treatment. Fibrogenesis is regulated by highly redundant multicellular signaling networks, potentially limiting the efficacy of single-target interventions. Moreover, advanced fibrosis remains partially self-sustaining and less reversible despite suppression of upstream profibrotic signaling. In preclinical MASH-fibrosis models, in vivo-generated CAR T cells, targeting fibroblast activation protein (FAP), selectively ablated activated hepatic stellate cells and myofibroblasts, thereby reversing extracellular matrix deposition and restoring liver homeostasis.75 This study provides a novel insight into an HSC-targeted clearance approach via reprogrammed adaptive immunity. Therefore, future antifibrotic strategies will likely require application within an optimized time windows, precision delivery systems, and rational combination approaches integrating metabolic, inflammatory, and fibrogenic pathway modulation.

Targeting Other Liver Cells: Treatments from Niche

Beyond hepatocytes, immune cells, and hepatic stellate cells, increasing evidence suggests that other hepatic niche populations, including liver sinusoidal endothelial cells (LSECs) and cholangiocytes, actively contribute to MASH progression through regulation of vascular homeostasis, inflammatory signaling, and microenvironmental remodeling. These cell populations, often receiving less therapeutic attention, emerge to play important roles in shaping fibrotic and inflammatory microenvironments, highlighted by recent advances in spatial and single-cell analyses.76 Therefore, targeting treatments from hepatic (ie, LSEC and cholangiocyte) niches represents a promising yet still largely experimental therapeutic direction.

LSECs play essential roles in maintaining hepatic microcirculation, substrate exchange, and immune tolerance through their highly fenestrated phenotype and angiocrine signaling functions.178,179 Importantly, ROCK2 activation drives phosphorylation of myosin light chain and actin cytoskeletal contraction in LSECs, promoting fenestration loss and capillarization. Capillarized LSECs subsequently shift from a vasoprotective to a profibrogenic phenotype, with enhanced TGF-β and PDGF signaling toward adjacent HSCs, thereby promoting HSC activation and collagen deposition.180,181 During MASH progression, LSECs facilitate capillarization characterized by loss of fenestrae, endothelial dysfunction, impaired nitric oxide signaling, and increased expression of pro-inflammatory and profibrotic mediators. These alterations promote immune cell recruitment, hepatocyte hypoxia, and HSC activation, thereby resulting in fibrosis progression and vascular remodeling.182 Therefore, therapeutic approaches targeting LSEC dysfunction have attracted increasing interest. Recent evidence highlights LSECs as a potentially direct therapeutic target for antifibrotic therapy, as semaglutide was reported to modulate intrahepatic endothelial Glucagon-like Peptide-1 (GLP-1) receptors to reverse sinusoidal stress and attenuate liver fibrosis independently of weight loss in preclinical models.183 Also, statins have demonstrated potential vascular-protective and antifibrotic effects beyond lipid lowering through improvement of endothelial nitric oxide bioavailability and suppression of inflammatory signaling.60,184,185 Furthermore, LSEC-targeting statin delivery alleviates capillarization and remodel stromal microenvironment in vivo.186,187 Additional experimental evidence was demonstrated in animal liver MASH models (ie, mouse, rat), highlighting anti-fibrotic188–190 and immunomodulatory191–194 effects via LSEC targeted approaches. In addition, cellular senescence likely contributes to LSEC dysfunction by promoting defenestration, capillarization, and loss of endothelial homeostatic signaling. Experimental studies indicate that senescent LSECs can disrupt endothelial zonation and promote hepatic steatosis, while age-associated LSEC dysfunction increases susceptibility to MASLD.195,196 However, the precise contribution of endothelial dysfunction to disease progression remains unclear, thereby restoration of LSEC phenotypes and functions alone to treat MASH still requires further research effort.

Cholangiocytes have also emerged as important contributors to fibrogenesis and chronic inflammatory remodeling in liver diseases.197 Under conditions of chronic injury, ductular reaction characterized by ductular reactive cell activation and expansion is particularly associated with advanced fibrosis and poor clinical outcomes, suggesting that biliary niche remodeling may actively participate in disease progression rather than merely representing a secondary injury response.198 Activated cholangiocytes secrete inflammatory cytokines, chemokines, and profibrotic mediators (also termed cholangiokines) that promote immune-cell recruitment and HSC activation. Beyond serving as passive markers of injury, reactive cholangiocytes can actively reshape the portal niche through secretion of chemokines, cytokines, extracellular vesicles (EVs), and profibrogenic mediators that coordinate macrophage recruitment and HSC activation. Moreover, persistent cholangiocyte stress and senescence may reinforce this response through a senescence-associated secretory phenotype, thereby sustaining ductular inflammation and fibrogenic remodeling. These findings suggest that therapeutic modulation of cholangiocyte activation states and their paracrine signaling, rather than complete suppression of ductular reaction, may represent a more physiologically appropriate strategy in advanced MASH.199–201 Our previous study demonstrates that inhibiting the acute phase protein orosomucoid 2 (ORM2) in intrahepatic cholangiocytes could ameliorate cell injury, macrophage inflammation and fibrogenesis, highlighting a potential protective approach for liver injury.202 Furthermore, developmental signaling pathways including Notch and Wnt signaling are critically involved in cholangiocyte activation and ductular reaction, prompting investigation of Notch inhibitors and Wnt pathway modulators as potential therapeutic strategies.203,204 Nevertheless, these pathways also regulate tissue regeneration, stem-cell maintenance, and epithelial homeostasis, raising concerns regarding off-target toxicity and impaired regenerative capacity following systemic inhibition.205 In addition, cholangiocyte senescence may also reinforce pathogenic biliary niche remodeling. Senescent cholangiocytes develop a SASP capable of recruiting macrophages and myofibroblasts, promoting TGF-β signaling, collagen deposition, and paracrine propagation of senescence to neighboring epithelial cells.206 Such a mechanism may be relevant to the ductular reaction and portal fibrotic remodeling observed in advanced MASH. Therefore, targeting cholangiocyte or ductular reaction may provide favorable effects on the amelioration of portal inflammation and fibrosis in cholangiopathies or conventional liver injury, remaining an alternative pathological modification approach for MASH.

Taken together, niche-targeted therapies directed at LSEC- and cholangiocyte-associated pathways remain at an early stage of development but provide an important conceptual expansion beyond traditional therapeutic models. Future progress will likely require improved understanding of spatial niche interactions, intercellular signaling dynamics, and disease-stage-specific functions.

Systemic and Multi-Cell Modulators: Orchestration of Pathological Networks

Even though cell-selective therapies offer mechanistic precision, the complex and interconnected nature of MASH pathogenesis suggests that interventions targeting multiple organs, pathways, and cell populations may provide broader therapeutic benefits. Indeed, most clinically advanced agents merely act on a single hepatic cell type but rather orchestrate systemic metabolic reprogramming, immunomodulation, and multi-cellular communication networks.8,207 Such therapies influence hepatic cells, immune cells, adipose tissue, and the gut-liver axis simultaneously, thereby interrupting the self-reinforcing cycles underlying inflammation and fibro