From Steatosis to Steatohepatitis: How Lipotoxicity, Inflammation, and Fibrogenesis Distinguish MASH from NAFL

Metabolic dysfunction-associated steatotic liver disease (MASLD) spans a spectrum from isolated steatosis (NAFL) to steatohepatitis with hepatocellular injury, inflammation, and fibrosis (MASH). Although both NAFL and MASH involve hepatic fat accumulation, progression to MASH reflects a qualitative shift: lipotoxic stress triggers organellar dysfunction and cell death; damage signals activate innate and adaptive immunity; and cross-talk with mesenchymal cells remodels the extracellular matrix, culminating in fibrosis. Recent studies (through 2025) refine this mechanistic framework and link it to measurable clinical phenotypes, noninvasive biomarkers, and emerging therapeutics.

Clinicopathologic distinctions and prognostic primacy of fibrosis

Histologically, MASH is defined by steatosis with hepatocellular ballooning and lobular inflammation, often accompanied by perisinusoidal (“chicken-wire”) fibrosis. By contrast, NAFL exhibits steatosis without significant injury or inflammation. This distinction is mechanistically grounded in lipotoxicity and innate immune activation rather than lipid quantity per se: system-wide insulin resistance drives steatosis, but lipotoxic species and immune activation separate MASH from simple fat accumulation 1. In a 2025 cholesterol-centric model, impaired hepatic cholesterol sensing via a dominant-negative LXRα mutation produced decreased triglycerides but increased unesterified cholesterol, rapidly provoking ballooning, immune infiltration, and fibrosis—underscoring that cholesterol accumulation, not triglycerides alone, is a critical MASH driver 2.

Fibrosis stage remains the strongest histologic predictor of outcomes. Across MASLD, fibrosis severity correlates with liver-related events and mortality, and noninvasive fibrosis staging has become central to clinical pathways. In type 2 diabetes (T2D) primary care cohorts, MASLD prevalence approaches 60%, with suspected advanced fibrosis in ~7%; obesity raises fibrotic MASLD risk eightfold 9. Notably, pediatric data show MASLD metabolic criteria identify children at higher risk of significant fibrosis than legacy NAFLD criteria (adjusted ~6-fold risk), validating the nomenclature change for risk detection 7. As MASH progresses to cirrhosis, sustained immune activation and matrix remodeling drive decompensation and hepatocellular carcinoma (HCC), including rising rates in non-cirrhotic MASLD—an emergent surveillance gap 3016.

Core pathobiology: lipotoxicity and organellar stress as the switch to MASH

The transition from NAFL to MASH is initiated by toxic lipid species—saturated free fatty acids (e.g., palmitate), ceramides, diacylglycerols, lysophosphatidylcholines, and free cholesterol—that overwhelm hepatocellular coping mechanisms. Multiple converging mechanisms have been elucidated:

Cholesterol-centric lipotoxicity

LXRα normally senses cholesterol derivatives to transcriptionally adjust lipid handling. Disabling cholesterol sensing (LXRα W441F) causes hepatic cholesterol overload, ballooning, immune infiltration, and fibrosis despite lower triglycerides, with a shift toward lipid-associated macrophages (LAMs; Trem2, Spp1 up) and loss of resident Kupffer cell markers (Clec4f). Synthetic LXR agonists reverse inflammation and fibrosis, directly linking cholesterol accumulation to steatohepatitis and its reversibility by restoring cholesterol homeostasis 2.

Mitochondrial dysfunction and inflammasome priming

CMPK2, a mitochondrial enzyme for DNA synthesis, is selectively upregulated in MASH (but not steatosis) in humans and mice. Hepatocyte-specific Cmpk2 deletion reduces injury, inflammation, and fibrosis while preserving steatosis, and pharmacologic inhibition dampens the NLRP3 inflammasome and pyroptosis—establishing mitochondrial bioenergetic stress and mtDNA-dependent signaling as critical to the steatosis-to-steatohepatitis transition 8. FFA-induced mitochondrial injury in hepatocytes releases mtDNA that activates macrophage AIM2 inflammasomes, triggering caspase‑1/gasdermin‑D (GSDMD) pyroptosis and IL‑1β/IL‑18 secretion; artemether attenuates this hepatocyte–macrophage axis 22.

ER stress as a pyroptosis gatekeeper

IRE‑1α, an ER stress sensor, directly stabilizes GSDMD to promote pyroptosis. Inhibiting IRE‑1α suppresses NLRP3 activation and GSDMD-driven cell death and ameliorates diet-induced MASH, identifying an ER stress–pyroptosis checkpoint 23.

Sphingolipid-driven pyroptosis

Free cholesterol upregulates sphingomyelin synthase 1 (SMS1), whose DAG product activates PKCδ and the NLRC4 inflammasome in hepatocytes to initiate caspase‑1/GSDMD pyroptosis. Pyroptotic hepatocytes release ATP, HMGB1, and mtDNA that trigger NLRP3 activation in Kupffer cells, amplifying IL‑1β and fibrosis. Knockdown of SMS1 prevents diet-induced steatohepatitis and fibrosis while steatosis persists—demonstrating that lipotoxicity-induced pyroptosis is both hepatocyte-intrinsic and a key upstream event that differentiates MASH from NAFL 21.

Autophagy defects and cholesterol accumulation

Chaperone-mediated autophagy (CMA) deficiency impairs HMGCR degradation, causing free cholesterol accumulation, ER stress, and hepatocyte injury. ER stress upregulates FOXM1, which further suppresses LAMP2A and CMA, creating a self-reinforcing loop that promotes inflammation and fibrosis; restoring CMA ameliorates MASH 24.

Apoptosis pathways

APP expression is dramatically increased in MASH hepatocytes across models and patients. APP interacts with death receptor 6 (DR6) to induce apoptotic signaling (cleaved caspase‑3/7), linking hepatocyte apoptosis to inflammation and fibrosis; APP suppression reduces MASH pathology 12. Epigenetic modulation also shapes susceptibility: inhibiting histone methyltransferase G9a increases palmitate-induced hepatocyte apoptosis and TGF‑β–driven stellate cell activation, exacerbating NASH 28.

Metabolic sensors and lipotoxicity modifiers

MKRN1 deficiency protects against steatosis, inflammation, and fibrosis in diet-induced MASH. Ebastine promotes MKRN1 self-ubiquitination and AMPK activation, reducing MASH features in mice 13. Sappanone A reduces hepatocyte lipotoxicity by upregulating Mup3, facilitating lipid transport and decreasing lipid deposition, inflammation, and fibrosis; Mup3 knockdown abrogates protection 14.

Collectively, these studies delineate a cascade in which lipotoxic species trigger mitochondrial and ER stress, activate caspase‑1/GSDMD pyroptosis and other death programs, and release DAMPs that recruit and activate immune cells—events that are absent or muted in simple steatosis.

Innate and adaptive immune activation as amplifiers of injury

Innate immunity is central to MASH. Kupffer cells and recruited monocyte-derived macrophages sense DAMPs and pathogen-associated signals (via TLRs and inflammasomes), propagate IL‑1β and TNF‑α–dominated inflammation, and facilitate fibrosis. LXRα-defective mice feature increased macrophage infiltration (CD68+), elevated Cx3cr1 and Tnf, and expansion of LAMs indicative of lipid-rich inflammatory niches 2. Pyroptotic DAMP release from hepatocytes activates macrophage NLRP3; macrophage-conditioned media then inflict hepatocyte injury, activate stellate cells, and promote collagen deposition—effects abolished when caspase‑1 is inhibited, directly linking macrophage NLRP3–caspase‑1 signaling to paracrine fibrogenesis 26.

Adaptive immunity contributes both inflammatory and pro-fibrotic cues. T cell subsets produce cytokines that perpetuate inflammation and fibrosis, and T cell–targeted therapies are entering clinical trials 3. Emerging evidence points to mast cell involvement: in MCD diet models, mast cell activation elevates TGF‑β1, driving stellate cell activation and collagen deposition; cannabigerol reduces mast cell infiltration and TGF‑β1, restraining fibrosis while steatosis persists 27. These immune changes underscore that the inflammatory milieu—not fat alone—defines MASH biology.

Stellate cell activation and fibrogenesis: the structural transition

Hepatic stellate cells (HSCs) are the principal fibrogenic effectors. A 2025 synthesis highlights how lipotoxic hepatocyte injury, macrophage-derived IL‑1β/TNF‑α, and TGF‑β signaling collectively drive HSC transdifferentiation with metabolic reprogramming (glycolysis, glutaminolysis) and myofibroblast differentiation (α‑SMA), culminating in collagen-rich extracellular matrix deposition 25. Liver sinusoidal endothelial cells (LSECs) undergo capillarization (loss of fenestrations), diminishing metabolic exchange and promoting pro-fibrotic signaling; mechanotransduction pathways (hedgehog, YAP/TAZ) further potentiate matrix remodeling. Genetic modifiers converge on these nodes: MBOAT7 rs641738 (loss-of-function) alters hepatocyte phospholipids to upregulate TAZ, inducing Indian hedgehog and accelerating fibrosis without worsening steatosis; restoring MBOAT7 slows fibrosis 33.

Systemic drivers and modifiers: insulin resistance, adipose and gut axes, and genetics

Systemic insulin resistance is the dominant driver of steatosis across MASLD 1; intriguingly, lean individuals with insulin resistance have significantly higher MASLD prevalence and steatosis burden, independent of BMI, and insulin resistance associates with advanced fibrosis 39. In T2D, MASLD is common with substantial fibrosis burden; visceral and subcutaneous abdominal adipose expansion and muscle fat infiltration accompany cardiac structural changes (reduced LV stroke volume and increased concentricity), affirming systemic consequences 9. The gut–liver axis contributes causally: intestinal TM6SF2 deficiency disrupts the barrier, enriches pathobionts, and elevates lysophosphatidic acid (LPA), which translocates to liver to promote steatohepatitis. Fecal transplantation transmits disease, and LPA receptor antagonism suppresses MASH, linking a genetic modifier to microbial and lipid mediators 31.

Genetics modulate susceptibility and progression. PNPLA3 I148M is the strongest common determinant of MASLD risk and progression to fibrosis; liver-directed PNPLA3 silencing reduced liver fat in phase 1 trials, with a histology-based phase 2b underway 6. In cross-sectional and longitudinal cohorts, PNPLA3 rs738409 minor allele carriage associates with higher ELF, FIB‑4, liver stiffness, and advanced fibrosis (ORs ~2.5–3.3), and combining FIB‑4 with PNPLA3/TM6SF2 genotypes stratifies 10‑year liver-related events from ~2% to ~54% across risk tiers 3834. Beyond hepatocyte-centric genes, cholesterol sensing (LXRα) and phospholipid remodeling (MBOAT7) exemplify how host genetics tune lipotoxic responses and fibrogenesis 233. Health disparities across sex, race/ethnicity, and socioeconomic status influence prevalence, diagnosis, and outcomes, emphasizing the need for culturally informed pathways 15.

Dynamic disease trajectory and feedback loops

The progression from steatosis to steatohepatitis and fibrosis involves interlocking feedbacks. Lipotoxicity provokes mitochondrial and ER stress; ER stress (IRE‑1α) and mitochondrial signals (CMPK2, mtDNA) gate pyroptosis (GSDMD) and amplify DAMP release; macrophage inflammasomes propagate IL‑1β and TNF‑α; paracrine TGF‑β and altered sinusoidal architecture activate HSCs; and nascent matrix stiffening activates hedgehog/YAP‑TAZ programs, reinforcing inflammation and fibrogenesis 2821232533. CMA failure and cholesterol accumulation create a self-perpetuating ER stress loop 24. Nonetheless, MASH remains modifiable: sustained weight loss reverses steatosis and steatohepatitis and can improve fibrosis; Roux-en-Y gastric bypass reduced BMI by ~8 points at 1 year and significantly lowered the Fibrotic NASH Index (FNI), with diagnostic performance supporting its use in monitoring 1710.

Clinical correlates and biomarkers linking mechanism to phenotype

Histology remains the gold standard for enrollment and endpoints, but reader variability constrains trials. A multisite validation of an AI-assisted system (AIM‑MASH) demonstrated superior accuracy versus unassisted reads for inflammation, ballooning, MAS ≥4 with ≥1 in each category, and MASH resolution, with non-inferior steatosis and fibrosis scoring—indicating AI may standardize histologic readouts in trials and practice 4. Mechanisms map onto evolving noninvasive biomarkers:

• Lipidomics distinguishes MASLD subtypes: among 481 quantified lipids, 210 differed across histologies, including 13 lipids linked to lobular inflammation, ballooning, and significant fibrosis. Dihexosylceramides emerged as novel fibrosis markers. A 14‑marker lipidomic panel predicted at‑risk MASH (AUROC 0.912 derivation; 0.76 validation) and advanced fibrosis (0.95), offering mechanistically grounded stratification 5.

• Multiparametric MRI: a two-step MRE-based stiffness followed by PDFF with corrected T1 (M‑PcT) outperformed FAST and MAST for at‑risk MASH (AUROC 0.832 vs 0.744 and 0.710), with higher PPV and NPV, positioning corrected T1 as an emerging activity-fibrosis biomarker 37.

• Fibrosis scores and algorithms: in health systems, a sequential FIB‑4 followed by ELF (cutoff 9.8) retained ~73% of patients in primary care, cutting hepatology referrals by ~26% and costs by ~8–9% compared with FIB‑4 alone; benefits persisted in higher BMI subgroups 40. Combining FIB‑4 with PNPLA3/TM6SF2 genotypes markedly sharpened prediction of liver-related events and mortality 34. AASLD-supported imaging letters reinforce adoption of noninvasive staging approaches 19. In T2D, transient elastography identified suspected advanced fibrosis in ~7% 9.

Therapeutic trials increasingly select endpoints reflecting mechanistic targets: hepatic fat fraction (HF‑MRI‑PDFF) reduction with FGF21 analog efruxifermin (65% reduction vs 10% with placebo on top of GLP‑1RA) accompanied improvements in injury and fibrosis markers 32, while corrected T1 and MRE can track inflammation and stiffness 37.

Therapeutic implications: targeting the nodes that distinguish MASH

The first MASH-specific drug approval—resmetirom, a selective thyroid hormone receptor‑β agonist—reflects an approach that enhances fatty acid disposal (lipophagy, β‑oxidation) and improves dyslipidemia with favorable tolerability, translating into histologic benefit 35. Clinical guidance affirms that only one medication has been approved to date, while weight loss via diet, exercise, and bariatric or endobariatric procedures remains the cornerstone of management 1117.

Mechanistic clarity is shaping combination and precision strategies. GLP‑1 and GIP receptor agonism shows causal protection against MASLD in Mendelian randomization—particularly GIPR signaling independent of BMI and diabetes—supporting incretin-based regimens beyond weight loss 36. FGF21 analogs added to GLP‑1RAs demonstrate synergistic steatosis reduction and biomarker improvement 32. Genetic precision is advancing: PNPLA3-targeted hepatic silencing reduces fat in 148M homozygotes, with histology-driven studies underway 6. Beyond metabolic correction, anti-inflammatory and anti-fibrotic strategies are being validated preclinically: caspase‑1 inhibitors that block macrophage NLRP3–caspase‑1 signaling prevent HSC activation 26; mast cell modulation (cannabigerol) reduces TGF‑β1 and fibrosis 27; restoring cholesterol sensing (LXR agonism) reverses fibrosis in cholesterol-driven models 2; and mitochondrial/ER stress checkpoints (CMPK2, IRE‑1α) attenuate pyroptosis and inflammation 823. Novel small molecules and repurposed agents that tune metabolic sensors (ebastine for MKRN1, AMPK activation) or reduce lipotoxicity (sappanone A via Mup3) also show preclinical promise 1314. A forward-looking view anticipates RNAi, mRNA, antibodies, PROTACs, and cell-based therapies delivered by advanced systems to target overlapping metabolic, inflammatory, and fibrotic pathways concurrently 41.

Table 1. NAFL versus MASH: clinicopathologic and biomarker contrasts

Table 2. Mechanistic nodes distinguishing MASH and therapeutic angles

Conclusion

MASH differs from NAFL less by the absolute amount of hepatic fat than by the hepatocellular response to lipid excess. Lipotoxic lipid species can induce mitochondrial and ER stress, trigger hepatocyte death pathways, and promote release of danger signals that recruit and activate hepatic immune cells; in parallel, sinusoidal dysfunction and paracrine signaling activate stellate cells and drive fibrogenesis. Genetic and systemic modifiers—including PNPLA3, MBOAT7, TM6SF2, impaired cholesterol handling, insulin resistance, and gut–liver axis dysfunction—shift this balance toward persistent inflammation and fibrosis. These insights are increasingly reflected in biomarker strategies that go beyond fat quantification, especially elastography-based fibrosis assessment, with cT1 and omics-based approaches still emerging. Therapeutically, weight loss remains foundational, while approved pharmacologic options now include resmetirom and semaglutide for selected patients; other approaches, including FGF21-based agents and more specific lipotoxic or inflammatory-fibrotic targets, remain investigational.