Genetic predeterminants and recent advancements in steatotic liver disease: A roadmap toward precision hepatology

Abstract

Steatotic liver disease (SLD) encompasses a group of disorders characterized by the excessive accumulation of fat in the liver. It is classified into four categories based on clinical manifestations: Metabolic dysfunction-associated SLD (MASLD), metabolic-alcohol-associated liver disease (ALD), ALD, and cryptogenic SLD. In the United States, its prevalence stands at 34.2%, making it the most common cause of cirrhosis and hepatocellular carcinoma (HCC). In addition to factors related to endocrine, nutrition, and medications, several genetic markers have been implicated in the disease's pathogenesis. Notable genes include PNPLA3, TM6SF2, GCKR, MBOAT7 and HSD17B13. These genetic polymorphisms can significantly impact prognosis and disease outcomes. For example, PNPLA3 is the most frequently associated gene with MASLD, increasing the risk of HCC by 12-fold and liver-related mortality by 18-fold. Furthermore, certain genetic markers are more prevalent in specific ethnic groups; for instance, PNPLA3 is common among Hispanics, while TM6SF2 is linked to higher fat content in African Americans. With a better understanding of the genetic factors involved in the pathogenesis of SLD, significant advancements have been made in diagnostics and therapeutics. This review explores the role of genetic factors in the disease's development, discusses current advancements in non-invasive diagnostic modalities, and examines therapeutic improvements based on these genetic insights to achieve better outcomes.

Key Words: Steatotic liver disease; PNPLA3; TM6SF2; GCKR; MBOAT7; HSD17B14; Metabolic dysfunction-associated steatotic liver disease; Hepatocellular carcinoma

Core Tip: Steatotic liver disease (SLD) is a chronic disorder, characterized by the excessive accumulation of fat in the liver. It is the most common cause of cirrhosis and hepatocellular carcinoma. Many genetic factors including PNPLA3, TM6SF2, GCKR, MBOAT7, HSD17B14 etc., contribute to the development of the disease apart from the various metabolic and endocrine etiologies. Deciphering these genetic markers will help not only in the understanding of pathogenesis but can lead to the development of the various targeted therapies and diagnostic modalities. In this article, we review the genetic markers associated with the SLD, their ethnic distribution and implication in the therapeutic and diagnostic advancements.

INTRODUCTION

Steatotic liver disease (SLD) affects over a third of the United States population, with an estimated prevalence of 34.2% (95%CI: 31.9%-36.5%) according to the data made available by the National Health and Nutrition Examination Survey over 4 years[1]. SLD refers to a group of diseases marked by excessive accumulation of stored fat in the liver, or hepatic steatosis. It is divided into four categories, based on its clinical manifestations, including: Metabolic dysfunction-associated SLD (MASLD), metabolic-alcohol-associated liver disease (ALD), ALD, and cryptogenic SLD (of uncertain etiology)[2].

The most common form, MASLD, was previously referred to as non-alcoholic fatty liver disease (NAFLD)[3]. MASLD is the world’s most prevalent cause of chronic liver disease. It is responsible for the highest liver-related mortality and morbidity, and it occurs in over 30% of the world’s population[4]. Its incidence varies across different regions, ranging from about 13.5% in Africa to 31.8% in the Middle East[5].

MASLD is the leading cause of cirrhosis and hepatocellular carcinoma (HCC)[5]. Cirrhosis marks the 11^th leading global cause of death & is responsible for 2%-4% of total deaths. Many associations have been identified with MASLD including endocrine disorders [polycystic ovarian syndrome (PCOS), hypothyroidism, hypopituitarism], viral hepatitis C (especially genotype 3), nutritional factors (total parenteral nutrition, malnutrition, rapid weight loss), genetic conditions (Wilson’s disease, celiac, lipodystrophies), and drugs (corticosteroids, methotrexate, amiodarone, tamoxifen, valproic acid, nucleoside reverse transcriptase inhibitors). Several genetic markers, including single-nucleotide polymorphisms (SNPs), have been linked to this condition. These comprise the PNPLA3 gene variant on the rs738409 allele, which is found in about 25% of the world’s population and exercises anywhere ranging from 10% to 75% among various population groups[6,7].

Other genetic variants such as TM6SF2, GCKR, MBOAT7, and HSD17B13 have shown strong associations with the onset and progression of MASLD. These SNPs influence crucial processes such as hepatic very low-density lipoprotein (VLDL) secretion, lipid droplet remodeling, and lipogenesis, advancing our understanding of the molecular mechanisms driving the disease[8].

This review explores the current understanding of genetic predispositions and their role in the development and progression of SLD, highlighting recent advancements and emerging insights into the molecular mechanisms driving the condition.

HISTORICAL CONTEXT AND EVOLUTION OF TERMINOLOGY

MASLD, previously known as NAFLD, is marked by excessive fat accumulation in hepatocytes, leading to progressive hepatic damage and dysfunction. It is a widespread chronic disease of the liver, impacting over 30% of the world’s population. The global prevalence with regional distribution of MASLD is illustrated in choropleth map in Figure 1.

Figure 1 Global prevalence of metabolic dysfunction-associated steatotic liver disease, based on the systematic review and meta-analysis by Younossi et al[14]. The map displays regional variation in metabolic dysfunction-associated steatotic liver disease prevalence, with the highest rates observed in Latin America (44.37%) and the Middle East and North Africa (MENA; 36.5%), followed by Southeast Asia (33.07%), North America (31.2%), and Australia (31.2%). Western Europe and East Asia show lower rates at 25.1% and 29.71%, respectively. These differences likely reflect a combination of genetic predisposition (e.g., PNPLA3 and TM6SF2 allele frequencies), dietary patterns, urbanization, and metabolic comorbidities[14].

It is a multidimensional condition influenced by the interconnection of metabolic and lifestyle factors[9].

The term "nonalcoholic steatohepatitis (NASH)", now rebranded as "metabolic dysfunction-associated steatohepatitis” (MASH), was first introduced in 1980 to describe a liver condition that mimicked the effects of excessive alcohol consumption in individuals who didn’t drink heavily and had no other notable liver diseases[10].

The term “nonalcoholic” was used to distinguish this condition from liver disease caused by excessive alcohol consumption. However, this term doesn’t provide a precise representation of its underlying pathogenesis. The term “nonalcoholic” might imply that people with alcohol-related liver disease are somehow more responsible for their condition, which can be stigmatizing. The term “fatty” might perpetuate negative stereotypes or shame associated with being overweight or obese[11]. Recent studies have shown that there may be shared biological processes between NAFLD and alcohol related liver disease, ALD, which could lead to new treatment approaches. The term NAFLD was too broad and encompassed many different conditions[11].

A worldwide initiative was undertaken to overhaul the NAFLD terminology, spearheaded by leading liver disease organizations through a structured consensus-building process. The working group pinpointed five crucial aspects to tackle: Rectifying the flaws in existing terminology, clarifying the significance of steatohepatitis, accounting for the influence of alcohol, evaluating the impact on public awareness and regulatory frameworks, and exploring whether a new designation could streamline research and treatment. This joint endeavor sought to establish a more precise and functional classification framework for fatty liver disease[12,13].

The consensus introduced a new terminology for fatty liver disease, with SLD serving as the umbrella term to encompass various causes of steatosis[12,13]. MASLD replaced NAFLD and now requires the presence of at least one cardiometabolic risk factor and hepatic steatosis. MASH replaced NASH. The term cryptogenic SLD is reserved for cases that don’t meet MASLD or alternative etiology criteria. The new nomenclature retained previous definitions for steatohepatitis and disease stages, and applied to advanced fibrosis/cirrhosis cases, even without steatosis[11].

GENETIC PREDETERMINANTS OF MASLD

Major genes involved in the progression of the disease

MASLD is a chronic liver condition with a high global prevalence of approximately 30%[14]. MASLD represents the initial stage of a liver disease spectrum that can progress to an inflammatory stage known as MASH. Over time, MASH can progress to fibrosis, cirrhosis, and eventually HCC. Importantly, MASLD is an increasingly recognized cause of morbidity and mortality and has become one of the leading indications for liver transplantation[15].

The progression of MASLD to cirrhosis is typically slow and influenced by comorbid conditions such as type 2 diabetes (T2D) mellitus, obesity, and specific genetic polymorphisms. The interplay between environmental factors and genetic predisposition contributes to the heterogeneous presentation of MASLD. Several studies have demonstrated the heritability of the disease. For instance, a multigenerational cohort study of Swedish adults found that first-degree relatives of patients with MASLD have an elevated relative risk of developing HCC, progressive liver diseases, and liver-related mortality[16].

Although the PNPLA3 gene is the most well-recognized genetic variant associated with MASLD susceptibility, other genes, such as the TM6SF2, GCKR, MBOAT7, and the HSD17B13 have been implicated in disease pathogenesis (Table 1)[17,18]. The substitution of methionine for isoleucine at codon 148 (I148M) in the PNPLA3 gene is the most common genetic variant involved in the pathogenesis of metabolic associated fatty liver disease (MAFLD). PNPLA3 has hydrolase activity against triglycerides and retinyl esters (Table 1). The 148M mutation results in the loss of hydrolase activity and buildup of triglycerides in the hepatocytes (Figure 2)[17,19]. It causes liver damage and prohibits the release of protein that can counteract liver fibrosis. The accumulation of 148M on the lipid droplet helps it escape degradation by the proteasomes and inhibits the activity of ATGL and other lipases. The variant promotes triglyceride accumulation by sequestering ABHD5, thereby limiting its availability to activate adipose triglyceride lipase. This mechanism increases the likelihood of disease progression to steatosis, steatohepatitis, cirrhosis, and ultimately HCC[20]. Several human and animal models have provided supporting evidence on the hepatic accumulation of fat with the mutated PNPLA3. Hence, the normal PNPLA3 is beneficial in preventing hepatic steatosis. This can serve as a potential therapeutic target[17].

Figure 2 Schematic representation of intracellular pathways highlighting the key genes PNPLA3, TM6SF2, GCKR, and GCKR involved in susceptibility to metabolic dysfunction-associated steatotic liver disease development. The schematic highlights how variants in PNPLA3 (rs738409), TM6SF2 (rs58542926), GCKR (rs641738C>T), and GCKR (rs1260326) disrupt lipid metabolism in hepatocytes. PNPLA3 impairs triglyceride hydrolysis, TM6SF2 impairs very low-density lipoprotein secretion, GCKR impairs phospholipid metabolism, while GCKR increases glucose uptake and lipogenesis. This figure was created by BioRender.com (Supplementary material).

Table 1 Genetic variants and their impact on steatotic liver disease.

Major genes	Variant	Functional impact of mutation	Outcome	Therapeutic target
PNPLA3	rs738409 (I148M)	Impairs triglyceride hydrolysis resulting in elevated hepatic triglyceride content	Hepatic steatosis	rs738409 (I148M)
			12 X risk of HCC
			18 X risk of liver-related mortality
TM6SF2	rs58542926	Impaired secretion of VLDL	Hepatic steatosis	E167K
			Progression of MASLD and MASH
GCKR	rs1260326 and rs780094	Disturbs glucokinase pathway	Elevated serum triglycerides	rs1260326
			Hepatic fibrosis
			Hepatic steatosis
GCKR	rs641738	Disrupts the phospholipid pathway by reducing levels of arachidonic acid-containing phosphatidylcholines	Increase in free polyunsaturated fatty acids	rs641738
			Increased hepatic fat content
			More severe liver damage
			Higher risk of fibrosis

HCC: Hepatocellular carcinoma; VLDL: Very low-density lipoprotein; MASLD: Metabolic dysfunction-associated steatotic liver disease; MASH: Metabolic dysfunction-associated steatohepatitis.

Inherited polymorphisms in these risk genes significantly impact disease outcomes. For example, the PNPLA3 variant can increase the risk of HCC up to 12-fold and liver-related mortality up to 18-fold[21,22].

Furthermore, a genome-wide association study (GWAS) revealed that the PNPLA3 (I148M) variant is most prevalent among Hispanic populations, with lower frequencies observed in European Americans and African Americans (Table 2). Other studies have also linked this allele to Asian populations, including those from India, China, and Malaysia[23,24].

Table 2 Ethnic predominance of major genes involved in metabolic dysfunction-associated steatotic liver disease.

Gene	Ethnic predominance
PNLPA3	Hispanics, Asians
TM6SF2	Caucasians and African Americans
GCKR	European Americans and Middle East
HSD17B13	Europeans

The loss-of-function variant of the TM6SF2 gene, which encodes the p. Glu167 Lys substitution, has been associated with hepatic steatosis[25]. This gene is expressed in the human liver and small intestine and plays a key role in regulating hepatic lipid metabolism. TM6SF2 plays a crucial role in the process that enhances triglyceride enrichment with apolipoprotein during the secretion of VLDL from the liver (Table 1 and Figure 2). The rs58542926 polymorphism causes a C>T change, resulting in an E to K substitution at position 167 (Figure 2). This modification leads to a loss of function, which increases triglyceride levels in the liver and decreases circulating lipoprotein levels. Moreover, TM6SF2 is also responsible for regulating lipid synthesis and the quantity of lipoprotein particles released (Table 1)[17]. Furthermore, TM6SF2 interacts with fatty acid-binding protein 5 to regulate fatty acid levels. Therefore, deletion or loss of function of TM6SF2 in the liver can impair the secretion of VLDL, leading to lipid accumulation and accelerating the progression of MASLD (Table 1)[26].

In a cohort of obese children and adolescents, the minor allele of the SNP rs58542926 in TM6SF2 was associated with pediatric MASLD. When stratified by ethnicity, this gene variant was linked to higher hepatic fat content in Caucasians and African Americans, elevated liver enzyme levels in Hispanics, and a more favorable lipoprotein profile in both Caucasians and Hispanics (Table 2)[27].

The GCKR gene encodes the glucokinase (GK) regulatory protein, which modulates and protects GK, an enzyme crucial for hepatic glucose metabolism. Two common variants of this gene, rs1260326 and rs780094, have been associated with a lower risk of T2D but a higher risk of MASLD. The GCKR, specifically rs780094C>T variant has been linked to more severe liver fibrosis and elevated serum triglyceride levels (Table 1)[28,29]. GCKR plays a crucial role in regulating glucose influx in hepatocytes, thereby controlling lipogenesis (Figure 2). A loss-of-function mutation in GCKR (rs1260326), which affects the P446 L, appears to contribute to increased hepatic fat accumulation. This mutation negatively impacts GK's response to fructose-6-phosphate, leading to continuous activation of hepatic glucose uptake. As a result, there is a decrease in fasting glucose and insulin levels, along with an increase in malonyl-CoA production. Malonyl-CoA serves as a substrate for enhanced fat production while simultaneously blocking fatty acid oxidation[17].

Another notable variant, rs641738C>T, located in the MBOAT7-TMC4 locus, has been recognized as a MASLD susceptibility gene (Figure 2). MBOAT7 is involved in the acyl-chain remodeling of phosphatidylinositol via the Land’s cycle, incorporating polyunsaturated fatty acyl chains into phosphatidylinositol. Loss-of-function variants in this gene result in reduced levels of arachidonic acid-containing phosphatidylcholines, an increase in free polyunsaturated fatty acids, and their pro-inflammatory metabolites (Table 1)[30]. This variant is associated with increased hepatic fat content, more severe liver damage, and a higher risk of fibrosis compared to individuals without the mutation[31]. The association between hepatic steatosis and this variant is more common in European Americans than in African or Hispanic Americans (Table 2)[31]. Although there is no knowledge in the literature about why different genetic variants are more prevalent in some ethnicities than the others. However, it can be attributed to the evolutionary selection, historical backgrounds such as diet and culture, climate, migration or genetic drift.

Additional genetic variants have been implicated in MASLD, including: NCAN, which encodes an adhesion molecule; PPP1R3B, which encodes a protein involved in glycogen breakdown; and HSD17B13, which is found in higher frequency among Europeans, encodes HSD17B13, a lipid droplet-associated protein involved in hepatic lipid metabolism (Table 2). These genes are associated with distinct alterations in serum and liver lipid profiles, as well as glycemic traits that can enhance genetic susceptibility to MASLD[32,33].

MASLD is a multifactorial disease resulting from the complex interplay between environmental influences and genetic predisposition. Understanding these genetic backgrounds holds significant potential for the development of genetic risk scores that allow better patient stratification, future gene-based therapies (e.g., CRISPR), and for achieving a more personalized treatment approach.

Influence of intracellular stresses on activation of genetic pathways leading to MASLD

Intracellular stress promotes the development and progression of MASLD by triggering genetic pathways. Hepatic lipid accumulation, a hallmark of MASLD, induces lipotoxicity, oxidative stress, and endoplasmic reticulum stress, which activate transcriptional programs that drive inflammation, fibrosis, and hepatocellular damage. These cellular stresses interact with inherited genetic variants, particularly in PNPLA3, TM6SF2, MBOAT7, GCKR, and HSD17B13, amplifying disease risk[26]. Intracellular stressors also modulate gene expression via epigenetic and transcriptional reprogramming, linking as shown in bioinformatic analyses that link progressive MASLD stages with altered expression of inflammation- and stress-related genes[34].

Immune system imbalance and genetic factors predisposing to MASLD

Immune system imbalance and genetic susceptibility interact closely in the initiation and progression of MASLD. Hepatic lipid accumulation triggers innate immune activation via Kupffer cells, monocyte-derived macrophages, and neutrophils, leading to the release of oxidized lipids, danger-associated molecular patterns, pro-inflammatory cytokines. These signals activate Kupffer cells through the TLR and NF-kB pathways, promoting the secretion of TNF-α, IL-1β, and TGF-β, which in turn amplify hepatic inflammation and drive stellate cell activation and fibrosis.

Genetic variants further modulate these immune responses[20,35]. Genetic variants such as PNPLA3 (I148M) exacerbate immune activation within the liver microenvironment; while TM6SF2 and MBOAT7 influence the hepatic immune response by altering lipid metabolism and membrane composition, which in turn affects immune cell signaling. MASLD involves both immune activation and failure in immunoregulatory pathways[35]. This interplay between immune dysregulation and genetic risk creates a pro-inflammatory hepatic environment that leads to disease progression.

Gene-gene and gene-environmental interactions

Genetic variants may interact additively or synergistically to influence hepatic fat accumulation, inflammation, and fibrosis. For example, combined carriage of risk alleles in PNPLA3 and TM6SF2 leads to more severe histological features than either variant alone[36]. In a 2020 study, researchers found that the combination of the PNPLA3, TM6SF2, and MBOAT7 genetic variants was associated with a higher risk of hepatic fat accumulation, inflammation, and fibrosis compared to their individual effects[37] Additionally, another study investigated the impact of PNPLA3, TM6SF2, GCKR, and MBOAT7 variants within the European population to assess the combined risk of different allele mutations. It was suggested that these combined mutations were linked to increased hepatic fat content and elevated transaminase levels. This evidence supports the idea of polygenic pathogenesis in MASLD, where the risk exponentially increases with each additional variant, as each genetic variant contributes differently to the disease process. The combination of various mutations results in a cumulative effect, leading to more severe disease outcomes[38]. MASLD is not solely genetically determined; instead, environmental exposures dynamically interact with genetic predisposition to modify disease courses. Furthermore, these genetic variants like PNPLA3, TM6SF2, GCKR, and MBOAT7 do not act independently but the modifiable lifestyle factors such as diet and physical activity seem to play a role. For instance, high intake of simple carbohydrates and saturated fats increases hepatic de novo lipogenesis (DNL), which synergizes with PNPLA3 and GCKR variants to worsen steatosis and inflammation[35,36]. In the human and animal studies, it was found that in subjects with homozygous I148M, a high fat or fructose diet or even a mixed diet led to increased hepatic steatosis and risk of MASLD. These models showed an increased inflammatory response with greater recruitment of Kupffer cells and macrophages leading to enhanced fibrosis. PNPLA3 mediated microbiota changes also seems to have played a role[39]. Genetic risk factors (GRS) involving various genetic variants are important predictors of fat accumulation in the liver, as well as inflammation and fibrosis. These genetic factors are significantly influenced by diet and lifestyle choices. For example, diets high in fats or fructose can increase the expression of the PNPLA3 gene, while following a Mediterranean diet has been shown to lessen its effects. These findings indicate that lifestyle changes, particularly regarding diet, can modify GRS, highlighting the importance of dietary strategies tailored to an individual’s genotype. Individuals with high-risk genotypes may need to make more significant dietary adjustments. Additionally, future models that combine GRS with environmental factors-such as diet quality, physical activity, and gut microbiota composition-could enhance risk assessment and lead to more personalized management of MASLD[40].

DIAGNOSTIC ADVANCEMENTS LINKED TO GENETICS

MASLD is a multifactorial disease with its pathophysiology heavily linked to genetic susceptibility, patients’ co-morbidities, diet, and lifestyle[41]. Liver biopsy is still considered the gold standard for the diagnosis of MASLD. However, it puts the patient at risk of infection. Other modalities such as ultrasound (US), computed tomography (CT), magnetic resonance imaging (MRI), fibro scan-aspartate aminotransferase scores and vibration controlled transient elastography are also employed to establish diagnosis[42,43]. However, with current understanding of its pathophysiology where genes such as PNPLA3, HSD17B13, and MBOAT7 have been identified to be major key players in its development, it has become imperative that genetic analysis must be included while establishing diagnosis of MASLD[44].

Role of genomics in non-invasive stratification

Considering the ever-increasing global burden of MASLD and its strong genetic association, the role of genomics is becoming heavily significant, especially when it comes to non-invasive risk stratification. Recent advancements in genomic technologies have facilitated the development of non-invasive methods to stratify the risk of MASLD in an individual. GWAS have identified genetic variants (e.g., PNPLA3, TM6SF2) correlating with MASLD susceptibility. This method has helped establish GRS tailored to liver-related complications[45]. Moreover, the role of epigenetic modifications cannot be overlooked as they equally influence disease progression and individual responses to treatment. An observational study that used a targeted multi-omics approach to examine the role of paroxenase-1 (PON1) in relation to obesity-associated fatty liver disease is a great example of how genomics helps stratify individuals at risk using non-invasive techniques[46]. This study integrated genetic variants, promoter methylation, expression profiles, and enzymatic activity of PON1. A cross-sectional study by Yang et al[44] highlights the critical role of genomics in non-invasive risk stratification of MASLD. They used MRI proton density fat fraction (MRI-PDFF) to quantify liver fat content (LFC) and correlated specific genetic variants such as PNPLA3, HSD17B13, MBOAT7, and UQCC1 with disease severity. This genetic profiling allows for a more precise assessment of MASLD risk without the need for invasive procedures like liver biopsy, enhancing early diagnosis and targeted interventions[44]. Wu et al[47] also showed how integrating genomics with other omics technologies such as proteomics, metabolomics and radiomics improves diagnostic techniques and yields comprehensive view of disease status and degree of fibrosis. While genomics offers promising avenues for non-invasive risk stratification in MASLD, challenges remain in the widespread clinical validation of these approaches. The integration of genetic insights with traditional diagnostic methods is essential for optimizing patient care and outcomes.

Imaging advancements with genetic correlation

Advancements in imaging play a pivotal role in diagnosing MASLD. They have significantly enhanced the understanding of genetic correlations that influence disease susceptibility and prognosis. Imaging techniques such as US, CT, controlled attenuation parameter (CAP) are employed for the diagnosis of MASLD[48]. However, MRI-PDFF has the highest diagnostic accuracy for the quantification of LFC[49]. Another cross-sectional study identified IncPRYP4-3, a long non-coding RNA, as a novel non-invasive biomarker for diagnosing and classifying subtypes of MASLD by using CAP imaging and serum samples[50]. The combination of imaging results with genetic data provides a comprehensive understanding of MASLD, potentially guiding therapeutic interventions.

Polygenic risk models

The implementation of GRS, particularly polygenic risk scores (PRS), in the stratification of patients with MASLD has emerged as a valuable tool for personalized medicine. These scores aggregate the effects of multiple genetic variants, notably those in PNPLA3, TM6SF2, and MBOAT7, which are consistently associated with disease progression, including fibrosis and HCC[51,52]. Polygenic risk models show promise in predicting the likelihood of MASLD in high-risk populations, especially when assessing diverse genetic backgrounds. PRS for MASLD has been linked to a significantly increased risk of severe liver disease, T2D, and HCC[53,54]. In a cohort study performed in United Kingdom, GWAS was performed to identify genetic variants associated with MASLD. Jamialahmadi et al[55] propose the use of partitioned PRS to delineate distinct biological pathways such as steatosis, inflammation, and fibrosis, thereby enabling better stratification and risk prediction. This layered approach can identify individuals at higher risk for advanced liver disease even in the absence of traditional metabolic risk factors. Similarly, De Vincentis et al[56] demonstrate that integrating PRS with clinical parameters enhances predictive accuracy, particularly in early identification of patients at risk of progression to cirrhosis or liver-related events. Sixteen significant SNPs were found associated with MASLD that differed from previously known NAFLD genetic profiles[57]. Another prospective cohort study showed that high PRS amplified the impact of MASLD on SLD and even extrahepatic outcomes[58]. Overall, incorporating genetic risk scores into MASLD management protocols allows for improved risk stratification, early intervention, and the tailoring of surveillance strategies. PRS based on the genes PNPLA3, TM6SF2, GCKR, and MBOAT7 have been well validated by various studies as predictors of hepatic fat accumulation, fibrosis, and HCC; their implementation in routine clinical hepatology is still limited and evolving. These scores provide additional prognostic value beyond conventional factors by identifying lean or metabolically healthy individuals who may be at higher risk. However, the clinical use of PRS is constrained by several practical scientific limitations, including a lack of standardization, modest effect sizes in population-wide models, and limited validation across diverse ethnic groups. Emerging studies indicate that PRS are most credible when integrated with environmental, metabolic, and imaging data to create multi-model risk assessments. Moving forward, prospective trials and real-world studies are necessary to determine whether genetic stratification can improve clinical outcomes or effectiveness in managing MASLD[40,59].

Artificial intelligence-assisted prediction

Diagnostic advancement has also progressed to artificial intelligence (AI) based prediction models which have demonstrated significant advantages over conventional methods in terms of sensitivity and specificity. These models make use of advanced algorithms to analyze various data types, leading to improved diagnostic accuracy. In one meta-analysis, pooled sensitivity rates of up to 97% were recorded by these AI models which signifies their ability to correctly identify patients with the disease[60]. A meta-analysis also showed similar results of around 94% specificity for fatty liver diagnosis[61]. Moreover, these models have been found to be comparable or superior in diagnostic accuracy to traditional methods with machine learning algorithms achieving similar or better results in predicting disease progression, especially when combined with electronic health records and imaging data. AI’s scalability allows for the rapid analysis of large datasets, making population-level screening feasible. Recurrent neural networks (RNNs) further enhance dynamic monitoring by processing time-series data to track disease progression over time. Deep learning models can provide quantitative measurements of liver fibrosis and steatosis, moving beyond subjective scoring. Additionally, AI systems excel at predicting future outcomes such as cirrhosis, HCC, and treatment responses, supporting effective risk stratification and personalized patient management[62,63].

While AI models show promising outcomes, Wong et al[62] discusses in depth the disadvantages of various AI approaches. While AI based models outshine due to their higher accuracy, non-invasiveness, dynamic monitoring and scalability, the use of AI for diagnosing MASLD comes with several other challenges. AI models are extremely data-reliant, requiring large, well-annotated and high-quality datasets, where biases or missing data can significantly impair performance. Deep learning models, for example convolutional neural networks (CNNs) and RNNs also demand difficult training and higher computational complexity. Recent advances in AI-assisted prediction for SLD focus on CNNs. Isshiki et al[64] used a transfer-learned VGG-16 CNN with echo-envelope parametric US images, achieving about 63% accuracy in grading steatosis. Meta-analyses from 2025 indicate that CNNs trained on B-mode US images achieve pooled area under the receiver operating characteristic curve ranging from 0.86 to 0.93 for classifying hepatic steatosis[65].

Generalizability remains one of the biggest concerns, as models trained in specific populations may lack performance across different ethnicities, lifestyles, or metabolic profiles. Where small sample sizes or high-dimensional data are involved, the high risk of overfitting remains. Many models have failed to discuss different populations like pediatrics, women, non-western population and lean patients. Clinical based models as compared to the imaging-based models have shown lower heterogeneity. Many AI models perform well within the studies but cannot replicate the same results in a clinical setting and hence, lack external validity. This raises serious questions about generalizability and fair selection of the study populations. In the global use of AI, the problem of ethnicity bias can arise with the human trained models in a particular setting[66]. The bias can arise from a number of factors such as temporal bias, institutional bias, selection bias, data bias, training bias, and financial bias[67]. Additionally, regulatory and ethical challenges complicate the deployment of AI in clinical settings requiring rigorous validation, certification, and compliance with health regulations. With the increased use of AI, ethical concerns like data ownership and healthcare privacy concerns can be raised. With the use of air and the traditional methods of obtaining consent can be obsoleted. Moreover, the liability issues in cases of adverse effects or unusual effects from the use of AI models can be encountered[66]. The theft of data is another challenging issue that can arise with the AI models[68]. Lastly, while RNNs offer some improvements, effectively modeling patient data that evolves over time remains an ongoing challenge[62,63].

THERAPEUTIC IMPLICATION OF GENETIC DISCOVERIES

Gene-specific drug targets (e.g., PNPLA3 inhibitors)

PNPLA3 has been identified as the strongest genetic factor for SLD. It is located on the long arm of chromosome 22, translates to a transmembrane protein which has triglyceride hydrolase activity, and plays a significant role in regulating lipid metabolism. Among all the variants known, PNPLA3 I148M is the largest contributor and has been linked to the development of hepatic steatosis, fibrosis, and inflammation, along with dysregulation in triglyceride hydrolysis, lipid droplet-Golgi trafficking, mitochondrial integrity, retinol metabolism, antioxidant defense mechanisms, and upregulation of TGF-β1 signaling, making it a promising target for treatment[69,70].

In addition to promoting lipid accumulation in the liver, overexpression of the PNPLA3 I148M has also been linked to exacerbation of oxidative stress, elevation in ceramide concentrations, and activation of the STAT3 signaling pathways, contributing to liver inflammation and fibrosis as seen in mice fed a sugar-enriched Western diet[69].

Based on this variant, many therapeutic strategies have emerged to target its pathological effects. One of these includes silencing the PNPLA3 (148M) at the RNA level using short hairpin RNA, small interfering RNA (siRNA), or antisense oligonucleotides (ASOs)[71]. Another study shows the use of GalNAc3-conjugated ASOs targeting PNPLA3 in a 148M knock-in mouse model led to a reduction in hepatitis fibrosis and steatosis, which highlights the potential of these therapeutics in reducing or eliminating liver damage[72]. To date, many ASO and siRNA-based therapies targeting PNPLA3 are under investigation. Among these, the AZD2693 is a GalNAc-conjugated ASO that has been designed to suppress the hepatic PNPLA3 mRNA, decreasing the formation of the PNPLA3 protein[69]. Another promising therapy is a GalNAc-siRNA-based candidate that has undergone a double-blind, randomized, and placebo-controlled phase 1 trial with both homozygous and heterozygous carriers of the PNPLA3 I148M risk allele, with early efficacy signals[73]. However, targeting PNPLA3's expression or function could affect other organs, such as the heart and blood vessels, possibly leading to adverse cardiovascular effects and other metabolic side effects. Research has indicated that inhibiting PNPLA3 may increase the risk of coronary artery disease and negatively impact cardiovascular outcomes[74,75]. Moreover, the long-term effects and safety of these therapies require further investigation. Currently, studies exploring specific inhibitors or activators of PNPLA3 are limited, highlighting the need for more research to fully understand the potential benefits and risks involved[74,75].

Precision medicine approaches

Recently, precision medicine approaches in the MASLD spectrum have been highlighted, which allows treatment based on lifestyle, genetics, and enhanced effectiveness with minimal side effects[76]. Precision medicine in MASLD focuses on the need to account for disease heterogeneity by considering key components such as sex, intestinal microbiota composition, endocrine disorders such as PCOS, physical activity levels, and reproductive status. By identifying patient subgroups based on these factors, clinicians can develop more targeted treatment strategies. Microbiota signatures, PRS, and predictive AI hold promise in guiding therapeutic decisions. A multi-component approach formed through these tools may increase the likelihood of achieving MASH resolution[77].

Nutrigenomics and personalized lifestyle interventions

Nutritional and lifestyle modifications play an important role in improving the outcomes of MASLD & its comorbidities. Nutritional genomics, an emerging field, explores how nutrients shape gene expression, genomic stability, and metabolic pathways, which contribute to personalized dietary strategies for the prevention of disease[78].

Diets rich in saturated fatty acids (SFAs) activate DNL and interfere with the hepatic lipid metabolism by activating PGC-1β, a coactivator of SREBP-1c, which is known to enhance expression of significant lipogenic genes like SCD-1, FAS, and DGAT[79]. SFA-rich diets have also been associated with increased endoplasmic reticulum stress, which may trigger unfolded protein response pathways and amplify caspase-3 activity, which contributes to elevated hepatocellular apoptosis[80]. Research in rodents has also shown that a low-carbohydrate, protein-rich diet can decrease the accumulation of adipose tissue, enhance regulation of glucose, and decrease steatosis, ultimately suppressing DNL[81]. Additionally, fructose-sweetened beverages have been linked to increased visceral fat, decreased insulin sensitivity, promote DNL, and play a role in hepatic lipid accumulation, which causes lipogenic gene expression and activity in the glycolytic pathway[82]. Minerals such as copper, when deficient and in combination with a high fructose diet, have been linked to worsening liver injury and promoting hepatic fat accumulation[83].

Future prospects of genetic therapies

The lack of approved therapies for MASLD results partly from a deficiency in human-relevant models for identifying and testing drug candidates. Recent advances have introduced human fetal hepatocyte organoids to duplicate early MAFLD changes, mainly steatosis, and drug screening using these cultured models has successfully identified compounds that reverse steatosis by suppressing DNL. Notably, the CRISPR-based FatTracer platform has been developed to screen genes that are involved in lipid metabolism and has highlighted FADS2 as a key regulator[84].

Emerging genomic analyses have highlighted therapeutic targets for MASLD-related fibrosis. In a trial by Lyu et al[85], increased expression of ATP-binding cassette transporter A1 (ABCA1) by exendin-4 decreases the hepatic lipid accumulation by CaMKK/CaMKIV/PREB pathway, and may potentially modulate the development of fatty liver disease. Many small-molecule drugs and biological agents targeting different pathways in MASH are currently in development, including some projects that have been discontinued. These therapies mainly modulate key cellular targets and molecular signals involved in inflammation, fibrosis, and lipid metabolism. Targets include receptors and enzymes such as DGAT2, SGLT-½, GLP-1R, PPARs, THR-β, FXR, ASK1, and CCR2/5, among others. Some hormone pathways, such as FGF19/21 and transcription factors like the CHREBP, are also being explored. These agents play a role in fatty acid oxidation, inflammatory signalling, and cholesterol metabolism, which eventually contribute to hepatic fibrosis[86].

Acetyl-CoA carboxylase inhibitors have shown promise in reducing hepatic injury, while the SCD inhibitor aramchol is pending a phase 3 trial for formulation refinement[87,88]. FASN-a key enzyme involved in the DNL pathway, has been linked to pro-inflammatory signaling, and denifanstat, a FASN inhibitor, is in its phase 2 trial in MASH patients, highlighting its therapeutic potential in dysregulated lipid metabolism[89]. Additionally, new clinical trials have shown that upregulation of FGF-21 decreases hepatic steatosis and improves the pathogenesis of MASH[90].

Another gene-based therapy involves adipose-derived mesenchymal stem cells (ADMSCs), which seem to be a promising option in regenerative medicine due to their immunomodulatory properties and differentiation into different cell types. Their contribution to MASLD stems from the release of adipokines, angiogenic factors, and growth factors[91]. ADMSC therapy has been shown to reduce fibrotic markers and pro-inflammatory cytokines in mice with metabolic syndrome and decrease disease progression via IL-17-mediated inflammation, which is linked to activation of hepatic stellate cells. Additionally, they may downregulate pro-inflammatory cytokines such as IL-4 and TNF-α while increasing the expression of anti-inflammatory cytokines in fibrosis induced by a high-fat diet[92,93]. Collectively, these advances in pharmacological, genetic, and cell-based therapies herald targeted and personalized treatment strategies in MASLD, improving clinical outcomes.

PROGNOSTIC AND CLINICAL OUTCOMES

MASLD is a chronic condition that occurs because of excessive fat accumulation in the liver, having close links with obesity, T2D, and circulatory comorbidities. The progression of MASLD is measured by the amount of scarring, also referred to as fibrosis, in the liver. Fatty liver at stage 0 means the deposition of fats in the liver with no damage or scarring. Stage 1 comes with mild fibrosis and minimal scarring. Stage 2 has moderate fibrosis, stage 3 is the advanced stage of fibrosis, and stage 4 is the stage at which extensive scarring occurs, liver function may be impaired, and significant cirrhosis occurs (Table 3)[92].

Table 3 Stages of metabolic associated fatty liver disease.

Stage	Brief description
Stage 0	Normal connective tissue of the liver with no scarring or fibrosis
Stage 1	Mild fibrosis
Stage 2	Moderate fibrosis
Stage 3	Advanced fibrosis
Stage 4	Cirrhosis

Chronic hepatic inflammation in MASH can trigger liver cancer development, even without cirrhosis. Hepatocarcinogenesis is a complex process involving multiple risk factors, including genomic instability, obesity, and diabetes. Metabolic changes, such as lipid and glucose metabolism alterations, contribute to the development of SLD. Additional factors, including genetic variants, stress, mitochondrial dysfunction, and immune response changes, drive disease progression to inflammation, scarring, and cancer[93].

HCC is the fourth main cause of cancer-related deaths globally. MASLD is the fastest-growing cause of HCC in many parts of the world, like the United States, China, and the United Kingdom[94-96].

MASLD was found to be the most common form of chronic liver disease, accounting for 59% of 4406 reported HCC cases in a large United States healthcare database study[97]. HCC has poor prognosis, with a 5-year survival rate of only 5%-15%. The growing prevalence of MASLD is accompanied by a rising incidence of MASLD-associated HCC, which increased by 9% annually in the United States between 2004 and 2009. MASLD-associated HCC has become the leading cause of HCC and the most rapidly growing indication for liver transplantation, with a significant increase from 2.1% to 16.2% between 2000 and 2016[95]. A dynamic Markov model predicts that the incidence of MASLD-associated HCC will increase by 122% by 2030 across eight countries. Many risk factors contribute to both the progression of MASLD to advanced fibrosis and the development of cancer, including old age, male sex, and advanced fibrosis or cirrhosis. Genetic variations, particularly those associated with PNPLA3, may contribute to MASLD progression and, in some cases, MASLD-associated HCC development. Cirrhosis is the most significant risk factor for HCC, increasing the risk by more than 10 times. MASLD-HCC can occur in earlier stages, with 25%-50% of cases developing in patients without cirrhosis, highlighting the complexity of this disease[95,98]. Genetic variations provide valuable insights into the risk of HCC. A PRS that includes variants in genes such as PNPLA3, TM6SF2, GCKR, and MBOAT7 is associated with an increased risk of HCC, even in individuals without fibrosis. Variants in the PNPLA3 gene are becoming increasingly important for predicting HCC risk and may have implications for managing patients awaiting transplantation. Integrating genetic information into comprehensive risk assessment models is beneficial for identifying high-risk individuals and guiding more personalized surveillance and prevention strategies[99]. However, the literature on whether these GRS can be used for decision-making on transplantation or surveillance is limited and needs further exploration.

MASLD is increasingly becoming a leading indication for liver transplantation globally. MASH cirrhosis accounts for about 10% of liver transplants in the US. The proportion of chronic liver disease attributed to MASLD has surged over time, increasing from nearly half to almost three-quarters of cases. As a result, MASLD’s representation among liver transplant recipients has grown. MASLD is currently the third most common reason for liver transplantation, following hepatitis C and alcohol-related liver disease. With MASLD’s prevalence expected to continue rising and hepatitis C treatments improving, MASLD may become the leading indication for liver transplants soon[100]. MASLD is expected to become the leading reason for liver transplants in the United States by 2030, with a high likelihood of disease recurrence after transplant. Patients who receive liver transplants for other reasons may develop new-onset MASLD due to the widespread presence of obesity and metabolic syndrome. Further research is needed to determine the best approach for managing MASLD in liver transplant patients[101]. After transplantation, graft steatosis can develop, either due to disease recurrence or de novo MASLD. Various factors contribute to graft steatosis, including metabolic syndrome, immunosuppressive medications, and genetics. Roughly 40% of individuals who receive a liver transplant may experience steatotic changes within 3-4 years after surgery, underscoring the prevalence of this issue[102,103].

The role of PNPLA3 gene polymorphisms in the development and progression of MASLD, MASH particularly in liver transplant patients, is very important. The PNPLA3 gene encodes adiponutrin, an enzyme involved in triglyceride metabolism. Certain genetic variations, such as the rs738409-G allele, are associated with increased LFC, advanced liver disease, and a higher risk of MASLD progression. Studies have shown that liver transplant recipients with the PNPLA3 risk allele are more likely to develop steatosis and metabolic complications, such as obesity and diabetes. Genetic screening for PNPLA3 polymorphisms may help identify high-risk patients and inform aggressive management strategies to prevent MASLD progression. IL28B genetic variations impact HCV treatment outcomes and potentially contribute to insulin resistance and fatty liver disease by acting on the interferon-stimulated genes via the JAK-STAT pathway[100]. IL28B is postulated to play a significant role in the post-transplant graft steatosis by increasing the risk of underlying metabolic syndrome. Furthermore, the risk of diabetes is significantly increased in those carrying the TT polymorphism of the IL28B gene[100]. However, there was no significant difference in the overall graft survival according to the recipient IL28B polymorphism.

CONCLUSION

As MASLD transitions from a traditionally diagnosed, lifestyle-modified disease to a genetically informed, precision-managed condition, it is imperative that future research bridges current gaps in data diversity, accessibility, and ethical governance. The convergence of genomics, AI, and personalized therapeutics holds the promise to not only mitigate disease burden but also revolutionize liver health globally. SLD, particularly in the context of metabolic dysfunction, is increasingly recognized as a condition with significant genetic predisposition. Key variants, including PNPLA3 I148M, TM6SF2, HSD17B13, MBOAT7 and GCKR, have been firmly established as major contributors to disease susceptibility, severity, and progression. These discoveries have reshaped our understanding of the disease beyond lifestyle and environmental factors, emphasizing the importance of inherited molecular pathways in hepatic fat accumulation, inflammation, and fibrosis. Advancements in genetic research have facilitated the development of non-invasive diagnostic tools, improved risk stratification, and laid the foundation for targeted treatment strategies. Emerging therapies such as ASOs and siRNA targeting pathogenic variants demonstrate promising potential to modify disease outcomes at the genetic level.

Despite these advances, many gaps remain. There is a need for broader, more diverse population studies to ensure findings are globally applicable, along with continued efforts to translate genetic discoveries into widely accessible clinical solutions. There is a need for extensive research involving multiethnic populations to understand how and why a specific genetic variant is more prevalent in certain groups. It's also important to determine if the current diagnostic models are free from biases. More real-world, multicenter studies are required to assess the clinical implications of the diagnostic and therapeutic approaches associated with these genetic variants. Additionally, an interesting point for discussion is the role of other genetic variants in developing therapeutic options for high-risk populations linked to that allele, as well as investigating their potential off-target effects and adverse reactions. The role of the GRS or PRS related to genetic variants should be explored for their potential in predicting HCC and guiding clinical decisions and outcomes in liver transplantation. It is also equally important to develop ethical framework to ensure data privacy in PRS development, provide equitable access to genetic testing for underserved populations, and prevent any algorithmic bias to translate precision hepatology from ideation to equitable clinical practice. As research continues to evolve, the integration of genetic insights into routine hepatology practice holds great promise for the early detection, prevention, and personalized management of SLD.