Molecular characteristics and carcinogenic mechanisms of aflatoxin B1-associated hepatocellular carcinoma: Recent advances

Abstract

Aflatoxin B1 (AFB1) is a Group-I chemical carcinogen. AFB1-associated hepatocellular carcinoma (AAHCC) shows distinct regional distribution worldwide with unique molecular profiles and carcinogenic mechanisms. AFB1 is metabolically activated by hepatic cytochrome P450 enzymes into reactive epoxides [aflatoxin B1-8,9-epoxide (AFBO)]. AFBO binds to DNA and form specific adducts. This process induces the characteristic G→T transversion and tumor protein p53 R249S mutation, serving as critical molecular fingerprints of AAHCC. AAHCC also presents typical mutational spectra, genomic structural aberrations, and epigenetic reprogramming. Malignant transformation is driven by p53 inactivation, abnormal activation of mitogen-activated protein kinase/Wnt pathways and remodeling of inflammatory immune microenvironment. These molecular characteristics differ markedly from non-AAHCC. Here, we systematically review the epidemiology, molecular phenotypes, core carcinogenic pathways, diagnostic biomarkers, and prevention strategies of AAHCC. We also analyze current research gaps and provide theoretical reference for precise prevention and clinical treatment.

Key Words: Aflatoxin B1-related hepatocellular carcinoma; Molecular characteristics; Carcinogenic mechanism; DNA adduct; Tumor protein p53 mutation; Epigenetics; Biomarker; Precision medicine

Core Tip: Aflatoxin B1 (AFB1) is a potent carcinogen characterized by hepatotropism, genotoxicity and carcinogenicity. AFB1-associated hepatocellular carcinoma (AAHCC) possesses distinct molecular profiles (including tumor protein p53 R249S mutation, C > A transversion, GCN motif preference, strand bias, and abnormal epigenetic modifications) and pathogenic mechanisms (including the formation of AFB1-exo-8,9-epoxide and AFB1-DNA adducts, abnormal DNA damage repair, and the accumulation of somatic mutation), exhibiting substantial heterogeneity from non-AAHCC.

INTRODUCTION

Hepatocellular carcinoma (HCC) is the most prevalent primary liver cancer globally, with striking regional disparities in incidence and mortality. High-burden areas are mainly distributed across sub-Saharan Africa, East and Southeast Asia[1]. Major acknowledged risk factors encompass chronic hepatitis B virus (HBV) infection, aflatoxin B1 (AFB1) exposure, alcoholic liver disease, non-alcoholic fatty liver disease, and metabolic syndrome[2]. According to distinct known etiologies, HCC is categorized into AFB1-associated HCC (AAHCC) and non-AAHCC. AAHCC refers to HCC with AFB1 exposure as the primary etiological factor or harboring AFB1-specific molecular alterations in tumor tissues. By contrast, non-AAHCC is defined as HCC predominantly driven by hepatitis virus infection, alcoholic liver disease, non-alcoholic fatty liver disease, and metabolic syndrome. The correlation between AFB1 and HCC was first discovered in the investigation of turkey X disease in the 1960s. Subsequent epidemiological surveys in high-risk areas of Africa and Asia confirmed its carcinogenic effect[3-5]. Classified as a group-I human carcinogen, AFB1 has driven intensive research on molecular signatures and pathogenesis of AAHCC.

AFB1 is metabolically activated by cytochrome p450 (CYP450) enzymes into AFB1-exo-8,9-epoxide (AFBO). It forms specific DNA adducts and triggers typical G→T transversion, predominantly manifested as tumor protein p53 (TP53) R249S mutation at codon 249[3,6]. The mutation frequency is closely correlated with AAHCC risk and geographical distribution[6-8]. High-resolution sequencing has characterized genome-wide mutational signatures induced by AFB1, such as C > A transversion, GCN motif preference and strand bias, facilitating the precise identification of AAHCC subgroups[9-11]. Recent evidence reveals that AAHCC pathogenesis involves multiple processes: Genomic mutation-induced genetic instability, nuclear factor kappa B activation and chronic inflammatory microenvironment triggered by oxidative stress, synergistic effects between AFB1 and hepatitis virus infection accelerating mutation accumulation, and elevated disease susceptibility arising from hereditary defects of DNA repair enzymes[1,3,6,12].

Despite remarkable advances in the molecular characteristics and pathogenesis of AAHCC, its detailed molecular mechanisms remain largely unclarified. Systematic exploration is insufficient concerning chronic low-dose AFB1 exposure, interactions with HBV, hepatitis C virus (HCV) and metabolic disorders, as well as epigenetic regulation. This review systematically summarizes updated research progress of AAHCC, focusing on three core facets: (1) Distinctive molecular signatures covering mutational spectra, genomic structural variants, and epigenetic aberrations; (2) Central pathogenic cascades including AFB1 metabolic activation, DNA adduct deposition, defective DNA damage repair, and inflammatory tumor microenvironment remodeling; and (3) Dysregulated signaling axes exemplified by p53 functional loss and overactivated mitogen-activated protein kinase (MAPK)/Wnt pathways. Combined with multi-omics data and clinical cohorts, it compares molecular heterogeneity between AAHCC and non-AAHCC, and evaluates advances in biomarker development, early screening, and primary prevention. Synthesizing multidisciplinary evidence from epidemiology, toxicology, genomics, and clinical medicine, this study deepens the mechanistic understanding of AAHCC. Also, it provides a theoretical basis and translational direction for precise intervention and individualized management based on molecular classification among high-risk populations (Figure 1).

Figure 1 Molecular characters of aflatoxin B1-associated hepatocellular carcinoma. Upon uptake of aflatoxin B1 (AFB1) by hepatocytes, cytochrome p450 (CYP450) enzymes mediate its metabolic conversion into the epoxide intermediate AFB1-exo-8,9-epoxide (AFBO). AFBO forms covalent cross-links with genomic DNA to produce DNA adducts; these lesions elicit targeted mutations in driver genes including TP53, alter global transcriptional landscapes, and induce epigenetic DNA dysregulation, collectively initiating AFB1-associated hepatocellular carcinoma (AAHCC). Beyond this primary mutagenic cascade, AFB1-triggered persistent inflammatory responses and genetic defects in DNA repair genes [e.g., xeroderma pigmentosum complementing group C (XPC), X-ray repair cross-complementing protein 4 (XRCC4)] serve as critical co-pathogenic drivers throughout AAHCC tumorigenesis. Furthermore, AFB1 exhibits synergistic oncogenic activity with hepatitis B virus (HBV) to expedite AAHCC disease progression. Taken together, the aforementioned molecular cascades define the unique molecular signature landscape that distinguishes AAHCC. Solid green arrows, representing metabolic reaction; Dashed green arrows, indicating correlations; Solid red arrows, indicating that AFBO is a critical intermediate during AFB1-induced AAHCC development; Dashed red arrows, indicating facilitative effects; Thin black lines, denoting synergistic or interactive effects with AFB1 in the progression of AAHCC; thick black lines, representing molecular events occurring in AAHCC.

EPIDEMIOLOGY

Sources of AFB1 and population exposure levels in different regions

AFB1 is a metabolic product of Aspergillus flavus and Aspergillus parasiticus. Humans are mainly exposed to AFB1 via intake of contaminated corn, peanuts, grains, and other crops[4]. Regional disparities in climate and dietary patterns lead to marked geographic variations in AFB1 exposure areas worldwide. High-exposure zones are tropical and subtropical regions with warm humid climate, predominantly sub-Saharan Africa, East and Southeast Asia. These high-exposure regions typically consist of developing countries where corn and peanuts are staple crops highly vulnerable to contamination. Furthermore, these areas suffer from inadequate grain storage, poor processing infrastructure, and a shortage of effective mold prevention and detoxification measures[1]. The average AFB1 concentration reaches 57.5 ± 6.8 μg/kg in peanuts and 29.7 ± 4.2 μg/kg in corn. Local children aged 1 to 6 receive AFB1 intakes of 350.0 ng/kg/day from peanuts and 146.4 ng/kg/day from corn, constituting a highly exposed population[13].

In Chiapas, Mexico, the seroprevalence of AFB1 among adults is 85.5% [95% confidence interval (CI): 72.1-93.1], with a median serum concentration of 0.117 pg/μL[8]. In Nigeria, children with severe acute malnutrition show a median AFB1-lysine albumin adduct level of 2.6 pg/mg albumin, and higher exposure is detected in stunted and wasted children[14]. Elevated AFB1 contamination in corn and peanuts is also found in Fusui, Guangxi, China, where the mean peripheral blood AFB1 adduct level in healthy residents is 2.18 Ln fmol/mg[15]. Cereals and derived products serve as the primary dietary exposure source in Chongqing, China, with estimated lifetime average daily dose ranging from 2.40 to 8.25 ng/kg bw/day and the 95^th percentile ranging from 9.51 ng/kg to 15.10 ng/kg bw/day[16].

In Karachi, Pakistan, 34% of patients with hepatitis C are exposed to AFB1, closely associated with high-risk dietary habits and insufficient health awareness[17]. Despite a broad dietary AFB1 exposure range of 0.25-35.0 ng/kg bw/day in Chile, the country maintains an extremely low HBV infection rate of 0.15% and low HCC incidence. This indicates viral hepatitis substantially modulates the carcinogenic potency of AFB1[18,19].

Notably, dietary components can modulate the toxic effects of AFB1. Both animal and human studies have demonstrated that diets low in sugar and protein tend to exacerbate aflatoxicosis, while high-protein diets may facilitate the in vivo detoxification of aflatoxins[4]. Furthermore, targeted interventions can effectively reduce AFB1 exposure. In high-exposure areas such as Qidong in China, shifting staple food from corn to rice together with strengthened food safety supervision has markedly decreased the levels of biomarkers including urinary AFB1-N-guanine and serum AFB1-lysine albumin adducts, accompanied by a decline in HCC mortality[5]. Randomized intervention trials have further verified that oltipraz, chlorophyll and broccoli sprout beverages can alleviate AFB1 burden by activating endogenous detoxification pathways[5].

Epidemiological distribution characteristics of AAHCC

AAHCC exhibits a global regional distribution, which coincides with areas of high AFB1 exposure and prevalent HBV infection[20-22]. Global spatial analysis reveals that regions with high AAHCC mortality largely overlap with high-risk areas of AFB1 exposure, and such correlation is more prominent under hot and humid climatic conditions[20]. In Gambia, AAHCC risk among children with HBV positivity caused by intake of contaminated peanuts and corn reaches 29 cases per 100000 population and 12 cases per 100000 population respectively[13]. In rural Kenya, approximately 1847 ± 514 HCC cases are annually attributed to AFB1-contaminated corn and 158 ± 52 cases to contaminated peanuts, totaling about 2000 ± 518 cases. The relevant risk presents a dose-dependent manner, with younger populations being more severely affected[23]. In Bangladesh, about 1311 annual HCC cases are attributable to dietary AFB1 exposure, accounting for 43.9% of the country’s total annual liver cancer cases[24].

Molecular epidemiological evidence further validates the causal association between AFB1 exposure and HCC. The TP53 R249S mutation (G-to-T transversion) at codon 249 is recognized as a typical molecular fingerprint of AFB1 exposure[25]. Among the 149 patients with HCC enrolled in Cameroon and the Central African Republic, the positive rate of R249S mutation was 24.8%, significantly higher than 5.6% in the control group[7]. In Guatemala, TP53 mutations were detected in 47% of patients with AAHCC, among whom 24% carried the R249S mutation. This mutation showed no overlap with HBV infection status, indicating that AFB1 can independently trigger HCC development even in regions with low HBV prevalence[26]. Genetic features linked to AFB1 exposure, including C > A transversions, GCN motif preference and strand bias, were identified in 9.8% of HCC specimens from Qidong, China, compared with merely 0.4%-3.5% in North America and Japan, and as high as 16% in Hong Kong[27,28]. Nevertheless, such a prominent correlation is not universally observed across all regions. A retrospective study involving 74 patients with HCC in Mexico City failed to detect any R249S mutation, suggesting that AFB1 exerts a limited role in local HCC pathogenesis, whose carcinogenic effect may be masked by other predominant risk factors such as metabolic syndrome[3].

There is a definite synergistic carcinogenic interaction between AFB1 exposure and HBV infection. Among patients with HBV-related HCC in southern China, the TP53 249Ser mutation induced by AFB1 exposure is correlated with elevated risk of tumor recurrence within 2 years after surgery [hazard ratio (HR) = 1.47, 95%CI: 1.02-2.18][29]. Nested case-control studies have also demonstrated that elevated levels of serum AFB1-albumin adducts markedly increase HCC risk in individuals without HBV/HCV infection and HCV-infected populations, with this effect being more pronounced in drinkers. AFB1 exposure serves as an independent predictive factor for HCV-associated HCC [adjusted odds ratio (OR) = 3.65, 95%CI: 1.32-10.10][30]. Whole-exome sequencing data confirm that AFB1 exposure is associated with inferior disease-specific survival (adjusted HR = 1.16, 95%CI: 1.01-1.33), and synergistic effects are observed in patients infected with HBV or HCV[31]. Furthermore, agricultural workers in Africa, especially those engaged in corn and peanut cultivation and processing, constitute a high-risk group due to occupational aflatoxin exposure[32]. Relevant prevention and control interventions including pre-harvest treatment and post-harvest management can reduce HCC risk by 23.0% to 95.1%[23]. In conclusion, the epidemiological distribution of AAHCC is determined not only by environmental exposure intensity but is also closely linked to host viral infection status, genetic susceptibility, and socioeconomic factors.

MOLECULAR FEATURES

Characteristic mutation spots and mutational signatures

AAHCC harbors distinct genomic mutational features. The most representative alteration is the TP53 R249S mutation at codon 249, characterized by AGG-to-AGT substitution leading to arginine-serine conversion. This typical G > T transversion arises from unrepaired DNA damage caused by AFB1-DNA adduct formation. Among 50 HCC tissue samples collected in Chiapas, Mexico, TP53 R249S mutation was detected in 3 cases (6.0%), and another 4 cases carried G > T transversions potentially induced by AFB1[8]. In 149 patients with HCC from Cameroon and the Central African Republic, the prevalence of TP53 R249S mutation is significantly higher than in controls[7]. This mutation is frequently documented in high AFB1 exposure areas including Qidong, China, and independently correlates with elevated tumor recurrence risk within 2 years after surgery (HR = 1.47, 95%CI: 1.02-2.18)[29].

Whole-genome sequencing has further uncovered mutational landscape characteristics of AAHCC beyond TP53. Among 49 patients with AAHCC in Qidong, China, ADGRB1 mutation frequently occurs apart from TP53 alteration and correlates with elevated tumor microvessel density[27]. Genome-wide analysis covering over 40000 mutations defines COSMIC mutational Signature 24 as the hallmark of AFB1 exposure. It is predominantly enriched with C > A transversions, preferentially arises at GCN sequence motifs, and presents remarkable strand bias[28]. This mutational signature has been validated in multiple experimental models and HCC specimens from high-exposure populations in China. In Hispanic residents of southern Texas, AFB1-specific mutations are identified in 7.3% (3/41) of HCC lesions. Such mutations are linked to younger onset age and unfavorable prognosis, and accumulate in pathways governing cell cycle regulation and BRCA1-dependent DNA damage response[33]. Analysis based on The Cancer Genome Atlas database demonstrates that AFB1 exposure load can be quantified via Signature 24, which is associated with poorer disease-specific survival (adjusted HR = 1.16; 95%CI: 1.01-1.33)[31].

Genomic structural variations and abnormal epigenetic modifications

Beyond point mutations, AFB1 exposure also triggers extensive aberrant epigenetic modifications including altered DNA methylation, dysregulated histone modification, abnormal non-coding RNA expression, and chromatin remodeling. Among these epigenetic alterations, DNA methylation represents the most crucial genomic and epigenetic abnormality characteristic of AAHCC. In vitro experiments reveal that human primary hepatocytes exposed to 0.3 μM AFB1 for 5 days sustain persistent genome-wide DNA methylation alterations and transcriptomic variations even 3 days after exposure cessation. Six hypomethylated and upregulated genes including thioredoxin reductase 1, proliferating cell nuclear antigen, cyclin K, diaphanous-related formin 3, Ras-related protein Rab-27A, and histone cluster 1 H2B family member f are closely implicated in oncogenic events, indicating that AFB1 can leave persistent epigenetic footprints[34]. In the AFB1-transformed human hepatocyte L02R cell model, hypermethylation of the tumor suppressor gene Runt-related transcription factor 3 (RUNX3) leads to its downregulation, a phenomenon detected in 70% (14/20) of clinical HCC specimens. Meanwhile, hypomethylation of LINE-1 repetitive elements and dynamic expression changes of DNA methyltransferases, TET demethylases and methyl-CpG binding protein MeCP2 are observed, further demonstrating sophisticated epigenetic dysregulation in AAHCC[35].

Animal model studies also support this conclusion. In offspring rats exposed to high-dose (5 mg/kg) or low-dose (0.5 mg/kg) AFB1 during gestation and lactation, DNA methylation levels of tumor suppressor and growth regulatory genes were markedly reduced in liver tissues, while elevated in peripheral blood (P < 0.05). This suggests that early AFB1 exposure raises adult HCC risk via tissue-specific epigenetic reprogramming[36]. Whole-exome sequencing reveals pronounced methylation heterogeneity even in genomically stable HCC tumors, implying epigenetic alterations exert pivotal roles in early tumor evolution[37].

Second, AFB1 broadly modulates core histone modifications, including histone acetylation and methylation[38-41]. Accumulating evidence has demonstrated that AFB1 significantly inhibits the activity of histone acetyltransferases while enhancing the expression and enzymatic activity of histone deacetylases in a dosedependent manner, thereby disrupting intracellular histone acetylation homeostasis[38]. In human cells, AFB1 exposure induces elevated H3K27 trimethylation (H3K27me3). This epigenetic alteration is catalyzed by EZH2, the catalytic subunit of the polycomb repressive complex 2 complex, which further promotes the recruitment of DNA methyltransferase 3 alpha (DNMT3a) to the p21 promoter and synergistically suppresses target gene transcription[39]. Notably, AFB1 exhibits distinct effects on histone modifications across different species and cell types. In porcine oocytes, AFB1 treatment reduces the levels of H3K27me3 and H3K4me2 but upregulates H3K9me3, indicating a sitespecific response of histone marks to AFB1 stress[40]. In Caenorhabditis elegans models, the F1 generation displays markedly increased H3K36me3 levels, which is closely associated with AFB1-induced transgenerational epigenetic toxicity[41]. Moreover, AFB1 enhances H2AK119 ubiquitination via upregulating the polycomb repressive proteins BMI-1 and EZH2, which further consolidates chromatin silencing[38]. Collectively, these AFB1-triggered histone aberrations promote chromatin condensation and transcriptional repression, preferentially affecting genes involved in cell cycle regulation and DNA repair.

Third, AFB1 exposure also induces widespread alterations in the expression profile of long non-coding RNAs (lncRNAs). In AFB1-induced rat HCC models, highly expressed lncRNAs in tumor tissues upregulate protein-coding genes involved in apoptosis modulation, DNA repair and cell cycle progression. By contrast, lncRNAs abundant in AFB1-resistant samples suppress these functional genes. It indicates that lncRNAs participate in the pathogenesis of AAHCC via modulating core gene regulatory networks[42]. Also, several studies have revealed that AFB1 induces dysregulated expression profiles of multiple microRNAs (miRNAs), such as miR-4651[43,44] and miR-24[45]. These aberrantly expressed miRNAs modulate the biological behaviors of HCC cells and affect the prognosis of patients with AAHCC. Mechanistically, these miRNAs may exert regulatory functions by targeting CYP450 2W1. Nevertheless, direct experimental evidence clarifying their definitive roles in the initiation and progression of AFB1-related HCC remains insufficient.

Finally, AFB1 exerts multifaceted regulatory effects on chromatin remodeling. On the one hand, AFB1 significantly upregulates the expression of BMI-1 and EZH2[38]. As the core catalytic enzyme for H3K27me3 modification, EZH2 not only mediates histone methylation directly but also recruits DNMT3a to the promoters of target genes such as p21. This establishes a synergistic silencing circuit coupled by histone modification and DNA methylation, thereby driving chromatin remodeling[39]. On the other hand, AFB1 indirectly modulates the higher-order chromatin structure through regulating polycomb group proteins[38]. Collectively, these alterations represent broad-spectrum chromatin remodeling events, which facilitate the formation of stable heterochromatin regions and profoundly influence the initiation and progression of AAHCC.

Comparison of molecular characteristics between AAHCC and non-AAHCC

AAHCC possesses distinctly different molecular profiles from non-AAHCC (Table 1). First, the oncogenic drivers of AAHCC differ from those of non-AAHCC[46-49]. The driver event underlying AFB1-induced hepatocellular carcinogenesis lies in the formation of covalent adducts between AFB1 and DNA, including 8,9-dihydro-8-(N⁷-guanyl)-9-hydroxy-AFB1 adduct (AFB1-GA) and ring-opened formamidopyrimidine AFB1 adduct. Accordingly, the generation of AFB1-DNA adducts in tumor tissue represents one of the most prominent molecular signatures of AAHCC.

Table 1 Different molecular characteristics between aflatoxin B1-associated hepatocellular carcinoma and non-aflatoxin B1-associated hepatocellular carcinoma.

Features	AAHCC	Non-AAHCC
Driven factor	The formation of AFB1-DNA adducts[46]	Chronic inflammation, metabolic reprogramming, etc.[47-49]
DNA adduct	AFB1-GA, AFB1-FAPYA[46]	Etheno-DNA adduct, DHP-DNA adduct, O6-methyl-G, etc.[52,53]
Gene mutation	TP53 R249S mutation (C>A transversion)[46]	TERT promoter mutations (ex. c.-124C>T, c.-146C>T) and CTNNB1 mutations[47]
Expression profile and signal pathway	Abnormal expression of genes involving in DNA repair and JAK-STAT[54]	Abnormal expression of genes involving in Wnt/β-catenin pathway[37]
DNA repair genes	Abnormal expression of XPC, XRCC4, XPD, etc.[46,55-57]	Abnormal expression of POLB[58]
Epigenetic modifications	Hypermethylation of RUNX3 and hypomethylation of TXNRD1, PCNA, CCNK, DIAPH3, RAB27A and HIST1H2BF[34]	HBx-mediated chromatin remodeling, hypermethylation of tumor suppressor genes, and elevated H3K27me3; histone acetylation (H3K18Ac/H3K27Ac)[59,60]

AAHCC: Aflatoxin B1-associated hepatocellular carcinoma; AFB1: Aflatoxin B1; CCNK: Cyclin K; DHP-DNA: Pyridylhydroxybutyl-DNA; DIAPH3: Diaphanous three; HIST1H2BF: Histone H2B type 1-C/E/F/G/I; JAK: Janus kinase: PCNA: Proliferating cell nuclear antigen; POLB: DNA polymerase beta; RUNX3: Runt-related transcription factor 3; STAT: Signal transducer and activator of transcription; TP53: Tumor protein p53; TXNRD1: Thioredoxin reductase 1; XPC: Xeroderma pigmentosum, complementation group C; XPD: Xeroderma pigmentosum, complementation group D; XRCC4: X-ray repair cross-complementing protein 4.

Second, different causes-related HCC subtypes exhibit markedly distinct mutational spectra. AAHCC is predominantly shaped by mutational Signature 24, which is defined by extensive C > A transversions and recurrent TP53 R249S mutations. By contrast, these molecular alterations are absent in patients with HCC residing in regions with low aflatoxin exposure, such as Mexico city[3], HCCs induced by other risk factors (viral hepatitis, steatohepatitis, metabolic syndrome) harbor characteristic driver mutations in the TERT promoter and CTNNB1 (encoding β-catenin) as primary molecular biomarkers[47,50,51-60] (Table 1).

Third, AAHCC exhibits a distinct transcriptomic signature of differentially expressed genes (DEGs). Integrated bioinformatic analysis of the GSE127791 and GSE64041 datasets uncovered 132 overlapping DEGs as well as 11 hub genes, among which suppressed expression of histidine-rich glycoprotein (HRG) and phosphoenolpyruvate carboxykinase 2, mitochondrial (PCK2) was correlated with unfavorable clinical outcomes[61]. A separate multi-omics investigation further identified six core genes whose expression is perturbed by AFB1. Specifically, RND3 (encodes Rho family GTPase 3) and PCK1 were profoundly downregulated, while Aurora kinase A (AURKA), branched-chain-amino-acid aminotransferase 2 (BCAT2), uridine-cytidine kinase 2 (UCK2) and CCNB1 displayed significant upregulation (P < 0.05). Subsequent molecular docking simulations confirmed robust and specific binding interactions between AFB1 and the proteins encoded by these candidate genes[62]. Furthermore, BUB1B and ribonucleotide reductase small subunit (RRM2) are overexpressed in AAHCC, with their expression strongly correlated with Janus kinase (JAK)- signal transducer and activator of transcription (STAT) pathway activation and immunotherapeutic efficacy[54].

Regarding pathway enrichment analysis, AAHCC tumors bearing AFB1-specific mutations display robust enrichment of gene sets governing cell cycle control and the BRCA1-dependent DNA damage response[33] and JAK-STAT pathway[54]. By contrast, non-AAHCC is primarily driven by dysregulation of the Wnt/β-catenin cascade, recurrent TERT promoter alterations, and mutations in chromatin remodelers including ARID1A[37]. It should be noted that AFB1 regulates the activation status of the Wnt/β-catenin pathway by inducing mutations in key genes of this cascade or remodeling the inflammatory microenvironment (see “KEY REGULATORY PATHWAYS AND EFFECTOR MOLECULES IN AAHCC PATHOGENESIS” section). Furthermore, AAHCC specimens consistently present an increased neoantigen load alongside upregulated PD-L1 expression[27], indicating heightened sensitivity to immune checkpoint blockade. This observation is consistent with abundant aryl hydrocarbon receptor (AHR) expression in AAHCC; AHR signaling further potentiates therapeutic responsiveness to anti-PD-L1 immunotherapy[63].

Fourth, AAHCC and non-AAHCC display prominent disparities in epigenetic landscapes. AAHCC is hallmarked by hypermethylation of the RUNX3 gene[34]. In contrast, non-AAHCC is dominated by HBx-driven chromatin remodeling, accompanied by hypermethylation of multiple tumor-suppressor genes, increased H3K27me3 levels, and altered histone acetylation marked by H3K18Ac and H3K27Ac[59,60].

Together, AAHCC possesses distinct molecular features across genomic mutation spectra, epigenetic modifications, transcriptional profiles and activated signaling pathways, clearly distinguishing it from HCC induced by other risk factors. Such molecular hallmarks support etiological origin tracing and offer promising clues for precise diagnosis and targeted treatment.

CORE PATHOGENIC MECHANISMS

Metabolic activation of AFB1 and formation of DNA adducts

AFB1 is the most hepatotoxic and carcinogenic member of the aflatoxin family. Its hepatocarcinogenic effect relies on metabolic activation in the liver. Catalyzed by hepatic CYP450 enzymes, mainly CYP1A2 and CYP3A4, AFB1 is converted into highly reactive AFBO[12,64]. This epoxide rapidly binds covalently to the N7 site of guanine in DNA, forming unstable AFB1-GA[65,66]. Under physiological conditions, the adduct undergoes ring opening and transforms into stable AFB1-FYPYA[66,67]. Evidence reveals that AFB1-FYPYA possesses potent mutagenicity, serving as the primary molecular basis for G→T transversion mutations[65,68].

AFB1-DNA adduct formation presents tissue specificity and predominantly occurs in the liver, attributed to abundant CYP450 expression. Adduct levels in organs like kidneys are merely about one hundredth of those in the liver[69]. In human primary hepatocytes, AFB1 is mostly metabolized into aflatoxin M (AFM), aflatoxin Q and aflatoxicol. Deficient pathways for glutathione conjugation and N-acetylcysteine derivative synthesis indicate humans have weaker AFB1 detoxification capacity compared with rodents[70]. AFB1-DNA adducts enhance the thermal stability of DNA double helix via 5’-end intercalation and base stacking, thereby interfering with normal DNA replication and repair[67].

Multiple high-sensitivity assays are available for quantitative detection of AFB1-DNA adducts. Liquid chromatography-tandem mass spectrometry combined with stable isotope internal standard can accurately detect 1.5 to 45 adducts per 10⁹ bases in mouse hepatic DNA[65,66]. The positive rate of serum AFB1-albumin adducts reaches 46% in human patients with HCC. Its concentration is positively correlated with bilirubin and adiponectin levels, and negatively correlated with albumin and LINE-1 methylation status, confirming the oncogenic role of AFB1-derived DNA adducts[71]. AFB1-DNA adducts are also identified in tumor and non-tumor liver tissues from low-exposure regions in Italy, suggesting long-term low-dose exposure may exert cumulative tumorigenic effects[72].

Abnormal DNA damage repair and accumulation of somatic mutations

Unrepaired AFB1-induced DNA adducts cause base mispairing during replication and lead to characteristic somatic mutations. The predominant alteration is G: C→T: A transversion, notably the TP53 R249S mutation at codon 249 (AGG→AGT), regarded as the molecular fingerprint of AFB1 exposure[5,10]. Around 10% of HCC specimens from high-exposure areas such as Qidong carry AFB1-specific mutational signatures featuring C > A transversions, GCN motif preference and strand bias, far exceeding the 0.4%-3.5% recorded in other regions[27]. The detection rate of R249S mutation hits about 25% in patients with HCC from Cameroon and the Central African Republic, markedly higher than in controls[7].

Cells eliminate AFB1-DNA adducts via multiple DNA repair pathways, mainly nucleotide excision repair (NER) and base excision repair. DNA glycosylase NEIL1 specifically recognizes and removes AFB1-FYPYA with verified catalytic activity in synthetic oligodeoxynucleotides and hepatic DNA of exposed mice[67,68]. NEIL1-deficient mice exhibit higher susceptibility to AAHCC, presenting elevated tumor incidence and larger average tumor size compared with wild-type mice, and the effect is more prominent than that in XPA mice with impaired NER function[73]. Polymorphic variants of NEIL1 such as A51V and G83D may abolish glycosylase activity, increasing hereditary susceptibility to AAHCC[74].

Repair efficiency is also modulated by the surrounding sequence context. The double-strand stability determined by 5’-flanking bases is negatively correlated with NEIL1-mediated repair capacity[67]. In repair-deficient models, loss of either NEIL1 or XPA yields AFB1-induced mutational profiles highly consistent with single-base substitution signature SBS24 in cancer somatic mutation datasets, with high cosine similarity[75]. Notably, XPA ablation disrupts the transcriptional strand bias of SBS24, whereas NEIL1 deficiency impairs adduct clearance in open chromatin regions, indicating complementary functions of distinct repair pathways across genomic domains[75].

Apart from direct mutagenesis, AFB1 facilitates mutation accumulation by disrupting DNA damage response pathways. It impairs ATM kinase activity, weakens G2/M checkpoint control, and triggers genomic instability along with compromised DNA double-strand break repair[64]. AFB1 activates the phosphoinositide 3-kinase (PI3K)/AKT pathway to sustain cell survival and trigger oncogene-induced senescence. It further overcomes senescence barriers via downregulating AT-rich interactive domain-containing protein 3A (ARID3A) and ARID3B, enabling damaged hepatocytes to proliferate incessantly[76]. These combined mechanisms drive progressive somatic mutation accumulation and eventual malignant transformation.

KEY REGULATORY PATHWAYS AND EFFECTOR MOLECULES IN AAHCC PATHOGENESIS

Inactivation mechanism of tumor suppressor genes such as TP53

AFB1 exposure induces characteristic mutations of TP53 in hepatocytes, among which R249S serves as a classic molecular marker derived from G→T transversion triggered by AFB1-DNA adducts. A Central African study reported that the prevalence of TP53 R249S mutation in patients with AAHCC is over four times higher than that in the control group. This mutation correlates with elevated AFP, high HBV viral load and abnormal blood cell counts[7]. No R249S mutation was detected in 74 patients with HCC in Mexico city, indicating limited relevance to aflatoxin exposure locally[3]. Beyond point mutation, AFB1 silences tumor suppressors via epigenetic regulation. In high-exposure regions like Qidong, promoter hypermethylation downregulates tissue inhibitor of metalloproteinase 3 expression, linked to aggravated tumor invasion and poor prognosis, revealing indirect impairment of tumor suppressive function by epigenetic repression[77].

Abnormal activation of oncogenic pathways including MAPK and Wnt/β-catenin

Accumulating studies have confirmed that AFB1 accelerates the malignant transformation of hepatocytes by activating the MAPK and Wnt/βcatenin signaling pathways. The detailed mechanisms by which AFB1 triggers MAPK activation are summarized as follows. First, AFB1 directly induces the phosphorylation of core kinases in the MAPK cascade. Both in vitro and in vivo models have demonstrated that AFB1 exposure markedly elevates the phosphorylation levels of key MAPK subfamilies, including ERK (MAPK1/3), JNK, and p38 MAPK[78-80]. Second, AFB1 indirectly activates MAPK signaling by modulating upstream receptors and signaling nodes. AFB1 can activate epidermal growth factor receptor (EGFR), thereby promoting downstream ERK phosphorylation. Increased expression of both EGFR and ERK has been observed after AFB1 exposure, indicating robust activation of the EGFR/ERK axis. Moreover, persistent AFB1 exposure alters the transcription of MAPK pathwayrelated genes during the malignant transformation of hepatic progenitor cells, which is closely associated with cell adhesion, motility, and signal transduction[79,81]. Third, AFB1 activates MAPK cascades synergistically via oxidative stress. AFB1triggered oxidative stress serves as a critical upstream stimulus for MAPK activation[78,82]. Fourth, AFB1 modulates MAPK activity through crosstalk with other signaling pathways. AFB1 simultaneously regulates multiple signaling cascades, including PI3K/AKT and Toll-like receptor/nuclear factor kappa B (NF-kB), which exhibit extensive crosstalk with the MAPK pathway[79,83]. Activated MAPK signaling not only mediates hepatocyte apoptosis and necrosis but also drives the malignant transformation of hepatic progenitor cells, constructs a protumor inflammatory microenvironment, and exacerbates oxidative damage and genomic instability, ultimately facilitating the initiation and progression of AAHCC.

Aberrant activation of the Wnt/β-catenin signaling pathway represents another critical molecular event driving AAHCC progression. The underlying activation mechanisms are summarized as follows. First, Wnt/β-catenin signaling is activated via genetic mutation-driven patterns. AAHCC frequently harbors somatic mutations in key regulatory genes of the Wnt cascade, including CTNNB1 and AXIN. Such mutations enhance β-catenin protein stability, promote cytoplasmic accumulation and subsequent nuclear translocation of β-catenin, and thereby induce persistent transcription of downstream target genes, leading to constitutive Wnt/β-catenin pathway activation[84,85]. Second, AFB1 indirectly triggers Wnt/β-catenin activation through microenvironmental regulation. Animal models have demonstrated that AFB1 exposure alters intestinal microbial composition, whereby Helicobacter hepaticus activates NF-κB-mediated inflammatory signaling and indirectly potentiates Wnt/β-catenin activity. This process is characterized by increased nuclear translocation of β-catenin and amplified downstream signaling in hepatocytes[86]. Activated Wnt/β-catenin signaling plays a pivotal role in AAHCC initiation, clonal evolution, and malignant phenotype determination, and has been recognized as a promising therapeutic target for AAHCC intervention[84,85,87]. Notably, emerging evidence indicates that AFB1 exerts context-dependent inhibitory effects on Wnt/β-catenin signaling via crosstalk with miRNA regulatory networks. In AFB1-treated human HCC cells, AFB1 upregulates miR-33a expression. Mechanistically, miR-33a-5p directly binds to the 3’-UTR of β-catenin mRNA, suppresses β-catenin expression, and ultimately attenuates Wnt/β-catenin pathway activity[88]. These findings suggest that the regulatory effects of AFB1 on Wnt/β-catenin signaling may be cell-type-dependent, dose-dependent, or context-specific, potentially leading to dual regulatory outcomes. Further in-depth investigations are warranted to clarify the bidirectional relationship between AFB1 exposure and Wnt/β-catenin pathway regulation in AAHCC.

Synergistic oncogenic effects of inflammatory microenvironment and dysregulated immunity

AFB1 promotes hepatocarcinogenesis not only by directly inducing DNA damage but also by triggering a chronic inflammatory microenvironment, which is tightly linked to AFB1-elicited oxidative stress. Although AFB1 itself is not a free radical, its intracellular metabolic activation substantially induces the massive accumulation of reactive oxygen species (ROS). The primary sources of ROS during AAHCC progression are summarized as follows. First, CYP450-mediated biotransformation of AFB1 perturbs the mitochondrial electron transport chain and facilitates mitochondrial ROS overproduction[89]. Second, AFB1 disrupts mitochondrial membrane potential and impairs the activity of respiratory chain complexes via dysregulating the expression of key mitochondrial genes [including NADH dehydrogenase subunit 1 protein (ND1)-ND6, cytochrome b, and cyclooxygenase-2], leading to electron leakage and further ROS elevation[90,91]. Third, AFB1 markedly depletes intracellular antioxidants and antioxidant enzymes, including glutathione, superoxide dismutase, catalase, glutathione peroxidase, and total antioxidant capacity, thereby compromising endogenous ROS scavenging capacity[89,92]. Fourth, excessive ROS attacks membrane lipids and induces the generation of toxic lipid peroxidation end products such as malondialdehyde, which further amplifies oxidative injury and exacerbates lipid peroxidation[89,92,93].

Oxidative stress and ROS induced by AFB1 activates the IKK-IκB-NF-κB pathway and triggers NF-κB nuclear translocation. Persistent activation sustains the expression of pro-inflammatory cytokines including interleukin 1 beta (IL-1β) and tumor necrosis factor alpha (TNF-α) as well as anti-apoptotic molecules, forming a tumor-prone inflammatory milieu[12]. In AFB1-induced rat HCC models, hepatic tissues exhibit elevated levels of pro-inflammatory cytokines TNF-α, IL-1α, and proliferating cell nuclear antigen, accompanied by decreased antioxidant enzyme activity and aggravated oxidative stress[94]. Patients with HCC from high AFB1 exposure areas present abundant mutation-derived neoantigens, massive lymphocyte infiltration and upregulated programmed death-ligand 1 (PD-L1), indicating activated yet immune checkpoint-suppressed tumor immune microenvironment[27]. AHR is highly expressed in AAHCC. It mediates AFB1 bioactivation and correlates with PD-L1 upregulation. CRISPR-Cas9 genome-wide screening identifies AHR as a pivotal regulator of AFB1 toxicity. HCC xenograft models with AHR overexpression achieve better therapeutic response to anti-PD-L1 treatment[63].

Notably, chronic inflammatory microenvironmental alterations and dysregulated immune responses triggered by hepatitis virus (e.g., HBV) infection synergistically promote the initiation and progression of AAHCC[95]. The underlying molecular mechanisms are elaborated as follows: (1) Persistent hepatic inflammation and aberrant immune signaling induce recurrent hepatocellular necrosis, compensatory proliferative regeneration and excessive oxidative stress, which exacerbates AFB1-mediated DNA damage in hepatocytes[95,96]; (2) Viral effector proteins (represented by HBx) suppress the NER pathway, impairing the clearance and repair of AFB1-DNA adducts and ultimately facilitating the accumulation of somatic mutations[95]; (3) A two-hit oncogenic mechanism driven by genic mutation and functional inhibition: AFB1-induced characteristic TP53 mutations, combined with HBx-mediated repression of p53 function, lead to nearly complete inactivation of the p53 signaling pathway under combined AFB1 and HBV exposure. This pathological change substantially enhances genomic instability and promotes clonal expansion of tumor cells[97,98]; and (4) HBV infection enhances the bioactivation of AFB1 via modulating the m⁶A RNA-binding protein YTHDF2 or activating the pregnane X receptor (PXR) signaling cascade, thereby amplifying the hepatotoxic and mutagenic effects of AFB1[99,100].

Recently, detection analyses of AAHCC specimens from regions with differential AFB1 exposure levels have identified aberrant expression of chromobox protein homolog 4[101-103], flotillin[104,105], and other key molecular regulators[15,106]. Such molecular dysregulation is tightly associated with tumor microenvironment remodeling and impairs the therapeutic responsiveness of targeted agents. Mechanistically, these molecules modulate core signaling cascades, including the mTOR, Notch, and VEGF pathways, through ubiquitination-mediated post-translational modification, thereby facilitating AAHCC malignant progression[103,104,107]. Collectively, these findings indicate that AFB1 exposure fosters a pro-inflammatory and immunosuppressive tumor microenvironment, which synergizes with genetic and epigenetic aberrations to drive AAHCC tumorigenesis and progression.

Collectively, these lines of evidence indicate that AFB1 can shape a pro-inflammatory and immunosuppressive microenvironment either alone or in synergy with other carcinogenic factors, and cooperate with genetic and epigenetic alterations to drive AAHCC progression. Nevertheless, the regulatory interactions between AFB1 and these target molecules, as well as their detailed molecular mechanisms, remain largely uncharacterized.

RESEARCH ADVANCES IN DIAGNOSIS AND PREVENTION

Advances in biomarker detection technologies for AFB1 exposure levels

Remarkable advances have been achieved in AFB1 exposure biomarker detection, mainly focusing on the quantitative analysis of specific adducts in blood and urine. Serum AFB1-lysine albumin adduct and urinary AFB1-GA are recognized as specific biomarkers for internal and biologically effective exposure dose, widely adopted to clarify the dose-response correlation between dietary aflatoxin intake and HCC[108]. High-performance liquid chromatography-tandem mass spectrometry combined with isotope dilution mass spectrometry enables precise quantification of AFB1-lysine in human specimens. Detected levels range from 0.2 pg/mg to 59.2 pg/mg albumin with a median of 2.6 pg/mg among children with severe acute malnutrition in Nigeria, and higher adduct concentrations are observed in children with stunting and kwashiorkor[14]. This technique also detects hepatic DNA adducts including AFB1-GA and AFB1-FYPYA enantiomers. Concentrations of 1.5 to 45 adducts per 10⁹ bases are detected in AFB1-treated mice, while none are found in controls, validating its application in early cancer risk assessment[66].

Non-traditional matrices such as hair have emerged as promising long-term exposure biomarkers beyond conventional biological fluids. Guinea pig experiments prove AFB1 can be detected in hair from day 3 to day 60 post exposure, supporting hair testing for chronic low-dose exposure assessment[109]. Urinary AFM1, a metabolite of AFB1 in vivo, serves as a short-term dietary exposure indicator. Despite potential cross-reactivity and matrix interference, commercial enzyme-linked immunosorbent assay remains applicable for preliminary epidemiological screening[110]. Urinary AFM1 levels show weak correlation with egg and dairy intake among Malaysian adults, verifying its biomarker reliability[111]. Multiple metabolites of AFB1 are detected in Ethiopian children’s urine via liquid chromatography-tandem mass spectrometry with a positive rate of 17%, confirming urinary biomarkers are effective for population exposure evaluation[112].

Early screening of AAHCC based on molecular characteristics

AAHCC possesses distinct molecular signatures, offering potential targets for early screening. The R249S mutation at codon 249 of TP53 is the most typical somatic mutation hallmark of aflatoxin exposure. In patients with AAHCC from Qidong, this mutation is accompanied by C > A transversion, GCN motif preference and strand bias, and correlates with elevated tumor vascular density and PD-L1 expression. Genomic analysis reveals 9.8% of Chinese HCC cases carry AFB1-specific mutations, markedly higher than the 0.4%-3.5% in other regions[27]. Digital droplet PCR detects the TP53 R249S mutation in 24.8% of patients with AAHCC in Cameroon and the Central African Republic, vs 5.6% in controls. Elevated mutant abundance is also found in circulating cell-free DNA, validating its use for early screening and exposure assessment in high-risk areas[7]. A 6.0% R249S mutation rate and multiple AFB1-associated G > T transversions are identified in AAHCC tissues from Chiapas, Mexico, confirming its value as a regional screening biomarker[8].

Omics studies have discovered novel biomarkers besides TP53 mutation. Integration of GSE127791 and GSE64041 datasets yields 132 DEGs specific to AAHCC. Downregulated HRG and PCK2 correlate with tumorigenesis and poor prognosis, serving promising early screening markers[61]. Upregulated AHR is closely linked to elevated PD-L1 in AAHCC. AHR knockout reduces adduct formation and improves cellular tolerance, making it a candidate target for screening and combined immunotherapy[63].

A case-control study conducted in regions with high AFB1 exposure explored the diagnostic value of serum miRNAs for early AAHCC. Serum miRNA-4651 level was markedly elevated in patients with AAHCC compared with non-AAHCC individuals, presenting superior diagnostic performance to AFP with the cutoff value of 400 ng/mL. The respective area under the curve values were 0.85 vs 0.72, and sensitivity stood at 78.1% vs 43.0%, with statistically significant differences. MiR-4651 also exhibits higher diagnostic efficacy for small and early-stage AAHCC relative to AFP[44]. Its expression level is correlated with AAHCC prognosis. For patients receiving post-operative adjuvant transarterial chemoembolization, elevated expression of miR-4651 in tumor tissues indicates favorable therapeutic response[43]. These findings suggest miR-4651 serves as a promising biomarker for early diagnosis and prognostic evaluation of AAHCC.

Advances in primary prevention strategies for AFB1 contamination control

Primary prevention is crucial for reducing the disease burden of aflatoxin-related HCC, covering source control, dietary intervention and policy supervision. Long-term practice in Qidong, Jiangsu serves as solid evidence. High corn contamination once caused high HCC incidence. Dietary shift to rice-based diversified food cut AFB1 exposure hundreds of times. The age-standardized HCC incidence dropped by 75% compared with the 1970s, with notable effects even among HBV-unvaccinated populations[113]. Clinical trials prove oltipraz, chlorophyllin and broccoli sprout juice can reduce AFB1-DNA and albumin adducts, validating their detoxification potential[6].

Most countries set total AFB1 limits of 4-20 ng/g for corn and peanuts. Yet existing standards fail to fully protect public health in low-income nations with high food consumption and prevalent hepatitis B, under the lifetime HCC risk threshold of 1 per 100000 population[114]. Despite enforced food regulations in Bangladesh, the risk reduction remains limited. Around 1311 annual HCC cases are attributable to AFB1 exposure, accounting for 43.9% of national liver cancer cases[24]. A comprehensive “5+1” prevention framework is proposed, covering source control, in-process detoxification, hierarchical management, short-term intervention, targeted follow-up for high-risk groups and climate-responsive early warning, to achieve systematic risk management[6].

Technologically, multiplex detection platforms based on nanomaterials and AI-integrated sensor arrays are developed for early aflatoxin warning. Combined with stage-specific biomarkers such as fungal spores, volatile organic compounds and regulatory genes, full-chain supervision from farm to table can be realized[115]. Biological control via antagonistic microorganisms to degrade aflatoxin has drawn wide attention, offering new solutions for food contamination prevention[116]. Severe contamination is found in peanuts (57.5 ± 6.8 mg/kg) and corn (29.7 ± 4.2 mg/kg) in Gambia. Children face daily exposure of 350.0 ng/kg and 146.4 ng/kg respectively, greatly elevating HCC risk, calling for urgent dietary intervention[13]. Multi-dimensional coordinated primary prevention plays an irreplaceable role in curbing AAHCC.

LIMITATIONS

Current studies investigating the molecular features and pathogenic mechanisms of AAHCC remain subject to several unresolved limitations. First, although International Agency for Research on Cancer has classified AFB1 as a group-I human carcinogen, and robust evidence confirms a causal association between high-level AFB1 exposure and HCC development, the oncogenic effects of chronic low-dose AFB1 exposure have yet to be understood. Most existing epidemiological and mechanistic research concentrates on populations residing in high-exposure areas, including Qidong and sub-Saharan Africa. Cumulative evidence demonstrates that even low-level AFB1 exposure is capable of inducing the characteristic TP53 R249S hotspot mutation[33] and facilitating tumor progression through persistent cumulative genetic injury[33,72]. Nevertheless, comprehensive evaluations regarding the long-term adverse health outcomes among populations with mild or low AFB1 exposure are still lacking. Furthermore, systematic prospective cohort studies remain scarce, which hinders precise quantification of the dose-response correlation between low-dose AFB1 exposure and HCC susceptibility[117,118].

Second, the synergistic carcinogenic mechanism between AFB1 and hepatitis viruses remains incompletely clarified. Quantitative risk assessment reveals that TP53 R249S mutation linked to AFB1 exposure significantly raises postoperative 2-year recurrence risk in patients with HBV-positive HCC (HR = 1.47, 95%CI: 1.02-2.18)[29]. Epidemiological surveys and TCGA data both demonstrate elevated AAHCC risk among individuals coinfected with HBV or HCV[31,119]. Co-exposure to AFB1 and HBV inactivates the FTCD-AS1-PXR-MASP1 axis, aggravating liver injury and malignant transformation[120]. Nevertheless, cross-ethnic and multicenter verification is lacking to confirm the universality of such mechanisms across diverse populations, viral loads and aflatoxin exposure doses. Studies on AFB1-HCV interaction are far less sufficient. One community cohort identifies high AFB-albumin adduct as an independent risk predictor of HCC among HCV carriers [adjusted odds ratio (OR) = 3.65, 95%CI: 1.32-10.10][30], while the underlying synergistic molecular pathways remain unclear.

The interaction of AFB1 with nutritional status, alcohol consumption and metabolic syndrome further complicates tumorigenesis. High-carbohydrate diets may exacerbate AFB1 toxicity, while high-protein diets potentially facilitate detoxification[4]. No TP53 R249S mutation is detected in non-cirrhotic HCC cases in Mexico city, with one third associated with steatohepatitis and metabolic syndrome, suggesting AFB1 effects can be masked by alternative pathogenic factors in non-endemic regions[3]. Reported immunotoxic effects of AFB1 show inconsistent impacts on specific immune markers across studies, hindering the interpretation of inflammation-mediated oncogenic synergy[121]. The intricate interaction network between AFB1 and diverse risk factors poses major research challenges. Integrated multi-omics analysis, longitudinal cohorts and cross-regional comparisons are urgently needed to clarify the exact synergistic carcinogenic mechanisms.

FUTURE RESEARCH DIRECTIONS AND PROSPECTS

Based on the current research progress in multi-omics molecular characteristic analysis and precise intervention of AAHCC, as well as the existing limitations and breakthrough opportunities in this field, future research can be carried out from two major dimensions: The optimization and upgrading of multi-omics technology systems, and the improvement of precise intervention strategies targeting molecular signatures. The detailed future research directions are elaborated as follows.

Optimized multi-omics strategies and in-depth mechanistic exploration

Current AAHCC multi-omics studies suffer from insufficient data standardization, obstacles in cross-platform data integration, and inadequate interpretation of biological mechanisms. Future research will focus on systematic technical improvement and precise mechanistic elucidation, covering three key aspects. First, it is essential to construct standardized and large-scale research cohorts and dynamic monitoring systems. Large prospective cohorts with graded AFB1 exposure levels and diverse HBV/HCV co-infection statuses should be established to overcome the limitations of existing datasets. Longitudinal multi-omics monitoring based on serial clinical samples will help precisely characterize the time-dose-effect relationship of AFB1-induced carcinogenesis and compensate for the deficiencies of static cross-sectional studies.

Second, multi-dimensional data fusion and full-chain mechanistic tracing should be achieved. Traditional multi-omics data, including genomics, transcriptomics, proteomics, metabolomics, and epigenomics, will be deeply integrated with single-cell sequencing and spatial transcriptomic data to dissect tumor microenvironment heterogeneity and regulatory patterns of immune cell infiltration during AAHCC progression. Combined with environmental exposure assessment techniques, such as quantitative detection of serum AFB1-lysine adducts, these approaches can reconstruct the full molecular trajectory from AFB1 environmental exposure and molecular perturbation to malignant hepatocyte transformation.

Third, optimized multi-omics algorithms are required to excavate novel biomarkers. Powered by machine learning and deep learning, unsupervised clustering and supervised learning models can be refined to identify high-sensitivity and high-specificity diagnostic biomarkers, prognostic scoring models, and therapeutic targets for AFB1-related HCC. Particular attention should be paid to AFB1-disturbed purine/pyrimidine metabolism, hexosamine signaling pathway, sialylation modification, and phosphorylated protein regulatory networks, so as to uncover novel oncogenic mechanisms underlying metabolic reprogramming and protein modification dysregulation.

Development and translation of precise intervention strategies based on specific molecular signatures

Leveraging the well-characterized AAHCC-specific genomic mutations, aberrant immune microenvironment, epigenetic disorders, and drug resistance-related molecular targets, future translational studies can be advanced in five directions: Early warning screening, targeted therapy, immune combination therapy, epigenetic intervention, and drug resistance reversal. First, develop novel techniques for early screening and dynamic monitoring of high-risk populations. Targeting AFB1-specific genomic alterations, including the TP53 R249S mutation and characteristic C > A mutational signature, optimized circulating tumor DNA liquid biopsy assays can be applied to establish a stratified early screening and dynamic surveillance system for HCC in AFB1 high-exposure regions, facilitating early diagnosis, early intervention, and precise risk stratification. Second, explore novel small-molecule targeted agents and individualized therapeutic regimens. High-specificity small-molecule inhibitors can be developed against core AFB1-regulated proteins, such as RND3, PCK1, AURKA, and BCAT2. Moreover, synthetic lethal strategies for HCC with high replication stress, represented by the combination of oxaliplatin and the WEE1 inhibitor adavosertib, deserve further validation to improve individualized targeted therapy systems for AAHCC. Third, innovate specific immune combination therapeutic strategies. Two AFB1-mediated immune regulatory axes can be prioritized for intervention: The AHR-PD-L1 axis to optimize anti-PD-L1 immunotherapy, and the IL-6-M2 macrophage regulatory axis to explore the synergistic efficacy of IL-6 inhibitors combined with PD-1 antibodies. Such combinatorial regimens can enhance intratumoral CD8+ T cell infiltration, reshape the anti-tumor immune microenvironment, and improve immunotherapeutic responses. Fourth, explore epigenetic-targeted intervention and preventive strategies. Given AFB1-induced global hepatic DNA hypomethylation and CCL20 promoter CpG site hypomethylation, the intervention effects of demethylase inhibitors and methyl donors need to be further validated. Notably, gestational AFB1 exposure confers high susceptibility to early-onset HCC in offspring, which highlights the urgent demand for identifying epigenetic targets for early prevention. Fifth, overcome clinical therapeutic resistance. CA2-mediated TACE resistance is a critical obstacle in AAHCC treatment. Future studies should verify the efficacy of restoring CA2 function and blocking AFB1-CA2 binding, aiming to reverse chemotherapy resistance and improve poor clinical prognosis of patients with AAHCC.

Core Directions for Clinical Translation and Implementation

The ultimate goal of future research is to promote the translational application of basic findings. Multi-omics-derived molecular biomarkers, diagnostic models, and intervention strategies can be translated into clinical practice to establish a precise diagnosis and treatment system based on AAHCC molecular subtyping. Furthermore, clinical trials of targeted therapy and combinatorial immunotherapy should be actively conducted in AFB1 high-prevalence regions to optimize clinical regimens, reduce regional health disparities, and substantially improve the overall prognosis of patients with AAHCC.

CONCLUSION

As a hepatotropic chemical carcinogen, AFB1 exerts a pivotal role in the initiation and progression of AAHCC. The disease presents distinctive regional distribution consistent with aflatoxin exposure patterns and unique molecular profiles[3,11]. Typical molecular traits cover signature mutational patterns featured by TP53 R249S mutation, C > A transversion, GCN motif preference and strand bias[11,27]; core gene regulatory networks with downregulated HRG, PCK2, RND3, PCK1 and upregulated BUB1B, RRM2, AURKA, CCNB1; and persistent DNA hypomethylation[34]. Dysregulated lncRNA expression and abnormal genes governing apoptosis, DNA repair and cell cycle also constitute vital alterations[34,42].

Mechanistically, AAHCC is closely associated with the genotoxicity of AFB1. It includes: (1) Metabolic activation of AFB1 generates DNA adducts and triggers characteristic genomic mutations[1,5]; (2) AFB1 induces oxidative stress and activates NF-κB signaling, fostering chronic inflammatory microenvironment to facilitate malignant transformation[12]; (3) Synergy between AFB1 and hepatitis viruses impairs DNA repair and accelerates mutation accumulation[2]; and (4) Hereditary deficiency of DNA repair enzymes weakens damage recovery capacity and elevates disease susceptibility[9].

Despite substantial advances in molecular pathology and pathogenesis of AAHCC, major challenges remain. The carcinogenic threshold of long-term low-dose exposure remains undefined, and interactive mechanisms between AFB1 and other risk factors require further exploration. Integrated multi-omics approaches are needed to dissect dynamic molecular networks triggered by AFB1. Precision screening tools and therapeutic strategies shall be developed targeting specific biomarkers including BUB1B, PCK2 and RND3. A clear exposure-molecule-clinical correlation exists in AAHCC. Elucidating its molecular signatures and pathogenic mechanisms carries great scientific and public health value for high-risk population screening, etiological classification and targeted prevention and treatment.

ACKNOWLEDGEMENTS

We thank Dr. Yuan-Feng Zhou for reference collection and management. All scientific analyses, discussions and manuscript writing of this review were independently completed by the authors.