Gastric cancer (GC) maintains its global dominance as the fifth most prevalent malignancy and third leading contributor to cancer mortality, with disproportionately high disease burden observed worldwide (1–7). While epidemiological trends show declining incidence in some geographical regions, delayed diagnosis persists due to nonspecific early clinical manifestations and inadequate screening biomarkers, culminating in a sobering 5-year survival rate below 30% for advanced-stage patients (8). The molecular pathogenesis of gastric cancer (GC) unfolds through Correa’s multi-step carcinogenic sequence. In this progression, Helicobacter pylori infection acts in synergy with chronic inflammatory processes and the accumulation of genetic/epigenetic aberrations, collectively driving the malignant transformation from gastritis to adenocarcinoma (9–12). Current therapeutic modalities, encompassing surgical resection, cytotoxic chemotherapy, and Human epidermal growth factor receptor 2 (HER2)-targeted agents, demonstrate limited effectiveness against tumor heterogeneity, metastatic dissemination, and therapy resistance mechanisms potentiated by the immunosuppressive tumor microenvironment (TME) (13–16). Even breakthrough immunotherapies exhibit modest clinical responses in GC (17–19), emphasizing the critical imperative to discover innovative therapeutic strategies targeting fundamental biological vulnerabilities such as metabolic reprogramming (20–22).

Metabolic reprogramming represents an essential adaptive mechanism enabling malignant cells to sustain uncontrolled proliferation under nutrient-constrained conditions (23–25). The Warburg effect-characterized by preferential glucose utilization through aerobic glycolysis despite oxygen availability – serves as a cornerstone of this metabolic rewiring in cancer biology (26–29). This bioenergetic shift facilitates rapid ATP generation while accumulating glycolytic intermediates for macromolecule biosynthesis, concurrently establishing an acidic, lactate-enriched TME that fosters immune escape, neoangiogenesis, and metastatic competence (30–35). Molecular orchestrators of this process include hypoxia-inducible factor-1α (HIF-1α), oncogenic kinase cascades, and rate-limiting glycolytic enzymes such as hexokinase 2 (HK2) and lactate dehydrogenase A (LDHA), all frequently overexpressed in malignant lesions (36–41). In GC pathogenesis, Hpylori-induced inflammatory signaling synergizes with oncogenic drivers to amplify glycolytic flux, establishing a self-reinforcing cycle that accelerates tumor progression and therapeutic resistance (42, 43). Preclinical investigations employing glycolytic pathway inhibitors-targeting glucose transporters (GLUTs), LDHA enzymatic activity, or lactate efflux mechanisms-have achieved significant suppression of tumor growth and chemotherapy desensitization, underscoring glycolysis inhibition as a promising therapeutic strategy (44–49).

The glycolytic phenotype exerts multifaceted impacts on gastric carcinogenesis and treatment responses. Gastric cancer (GC) cells display a striking reliance on glycolysis, driven by constitutive activation of the PI3K/AKT/mTOR signaling pathway and stabilization of HIF-1α—effects often amplified by Helicobacter pylori-associated chronic inflammation (50–56). This metabolic adaptation not only fuels unchecked proliferation but also generates an immunosuppressive, pro-metastatic niche through extracellular acidification and lactate accumulation (57–59). Clinically relevant glycolytic markers including HK2 and LDHA demonstrate strong correlations with advanced tumor stage, chemotherapy failure, and poor prognosis. Mechanistically, LDHA-generated lactate enhances β-catenin pathway activation, promoting cancer stem cell maintenance (60, 61). Emerging evidence reveals metabolic heterogeneity across GC molecular subtypes, presenting both challenges and opportunities for precision targeting. While preclinical models demonstrate encouraging antitumor effects with glycolytic inhibitors, clinical translation remains hampered by limited GC-specific trials and incomplete understanding of lactate’s dual metabolic/signaling roles. This comprehensive review analyzes the pathophysiological significance of glycolytic remodeling in GC and evaluates innovative therapeutic approaches, including metabolic inhibitor-immunotherapy combinations and nanoparticle-mediated drug delivery systems, that may overcome current limitations in targeted therapy development.

Gastric cancer (GC) progression is intricately tied to metabolic reprogramming, with dysregulated glycolytic flux—manifested as the Warburg effect—serving as a hallmark of tumor bioenergetics. Emerging evidence underscores the pivotal role of epigenetic mechanisms in orchestrating this metabolic shift through multilayered regulatory networks. Non-coding RNAs, including circRNAs and lncRNAs, dictate glycolytic adaptation via miRNA sponging, RNA splicing modulation, and epigenetic remodeling, while viral miRNAs further disrupt host metabolic checkpoints. Concurrently, dynamic m6A epitranscriptomic modifications fine-tune glycolytic enzyme expression through methyltransferase/eraser-mediated RNA methylation cycles. Post-translational modifications of metabolic kinases and transporters add another regulatory stratum, directly modulating enzyme stability and activity. These interconnected mechanisms collectively sustain metabolic plasticity, drive therapeutic resistance, and establish tumor-microenvironment crosstalk. The complexity of this epigenetic-metabolic interplay is systematically dissected in subsequent sections, with comprehensive regulatory hierarchies and molecular interactions detailed in Figure 3 and Table 1 . Elucidating these pathways unveils novel vulnerabilities for precision therapeutic targeting in GC.

Metabolic enzyme networks constitute actionable therapeutic targets in GC, driving tumor progression through isoform-specific regulation, dynamic structural modulation, and auxiliary pathway exploitation. Hexokinase isoforms (HK1, HK2, HKDC1) exhibit divergent oncogenic mechanisms, from mitochondrial-nuclear crosstalk to chemoresistance-linked redox cycling. PKM2 functions as a pleiotropic metabolic rheostat, integrating glycolytic flux with proliferative and inflammatory signaling via phosphorylation-dependent oligomerization and ubiquitination dynamics. Beyond canonical enzymes, auxiliary players such as PGAM1 and ALDOB expand targetable space by coupling metabolic rewiring with immune evasion and DNA damage response. These networks are exploitable through isoform-specific inhibitors, allosteric modulators, and RNA-based therapeutics, supported by advanced metabolic imaging for real-time therapeutic monitoring. The molecular hierarchies and therapeutic strategies governing these enzyme networks are systematically cataloged in Table 2 , providing a roadmap for precision targeting of GC metabolic vulnerabilities.

Metabolic signaling networks in gastric cancer (GC) function as dynamic regulatory hubs, integrating oncogenic drivers with bioenergetic reprogramming to fuel malignant progression and therapy resistance. Core pathways—including the AMPK/mTOR axis, hypoxic signaling, β-catenin cascades, Notch circuits, and Hippo/YAP effectors—exhibit bidirectional crosstalk with metabolic enzymes, enabling adaptive survival under fluctuating nutrient availability and therapeutic pressure. These networks operate through multilayered mechanisms: epitranscriptomic RNA methylation, nutrient-sensitive protein stabilization, and subtype-specific epigenetic rewiring, which collectively synchronize glycolytic flux with programs governing proliferation, stemness, and immune evasion. Therapeutic vulnerabilities emerge from pathway interdependencies, as exemplified by combinatorial targeting of metabolic-transcriptional feedback loops and spatial regulation of enzyme-transporter complexes. The molecular architecture and hierarchical interactions of these signaling-metabolic axes are systematically mapped in Figure 4 , with key regulatory nodes and therapeutic strategies cataloged in Table 3 . Deciphering this circuitry provides a framework for precision interventions that disrupt metabolic adaptation while counteracting compensatory oncogenic signaling.

The intricate crosstalk between metabolic reprogramming and oncogenic signaling in GC drives therapeutic resistance and malignant progression through multilayered regulatory circuits. Central to this interplay is the AMPK/mTOR axis, where the FTO-PRKAA1 axis (121) stabilizes AMPK’s catalytic subunit PRKAA1 via m6A demethylation, paradoxically activating glycolytic flux while maintaining redox homeostasis under nutrient deprivation. This pro-survival adaptation diverges from AMPK’s classical energy-sensing role, implicating non-canonical HIF1α-mediated signaling. Simultaneously, the RNA-binding protein PUM1 (52) stabilizes DEPTOR mRNA to suppress mTORC1 activity, yet activates compensatory PI3K-Akt signaling—a dual mechanism enabling tumor proliferation during metabolic stress. The Skp2-AKT/mTOR axis (141) further integrates growth factor signaling with glucose metabolism, where Skp2 enhances GLUT1-mediated uptake and LDHA activation, creating a self-reinforcing glucose-lactate cycle. Preclinical validation using thioridazine, an antipsychotic that degrades Skp2, demonstrates synergistic suppression of HER2⁺ GC growth in patient-derived xenograft (PDX) models when combined with trastuzumab, underscoring the therapeutic value of targeting metabolic-growth pathway intersections.

Hypoxic signaling pathways shape GC metabolism through bidirectional epigenetic-metabolic coupling. METTL14-mediated m6A methylation (119) silences the tumor suppressor LHPP, releasing GSK3β to degrade HIF1α and attenuate glycolysis. Conversely, hypoxia-induced HIF1α (142) transcriptionally activates chemokine C-C Motif Chemokine Ligand 7(CCL7), establishing a HIF1α-CCL7-KIAA1199 feedforward loop that amplifies glycolytic output and EMT. Genetic ablation of CCL7 disrupts hypoxia-driven cancer stemness, revealing its critical role in metabolic plasticity. Therapeutic strategies combining LHPP agonists, CCL7-neutralizing antibodies, and m6A writer inhibitors show promise in preclinical models, with single-cell spatial transcriptomics emerging as a tool to map hypoxic niche-specific vulnerabilities.

The β-catenin pathway serves as a metabolic-proliferative nexus in GC. Lysophosphatidic acid (LPA) signaling through LPAR2 (143) activates β-catenin nuclear translocation, upregulating c-MYC and Cyclin D1 while coordinating mitochondrial OXPHOS and glycolysis for rapid ATP generation. Pharmacological LPAR2 antagonists concurrently inhibit β-catenin signaling and metabolic activation, demonstrating dual therapeutic efficacy. Spatial regulation is exemplified by PDLIM1 (130), which anchors HK2 at mitochondria to stabilize β-catenin by physically blocking GSK3β-mediated degradation—a mechanism disrupted under glucose deprivation, revealing nutrient-sensitive decoupling of metabolic and proliferative signaling. Combinatorial targeting of LPAR2 and PDLIM1 using small-molecule inhibitors, alongside CRISPR/dCas9-mediated epigenetic editing of β-catenin target genes, may synergistically suppress β-catenin-driven malignancy.

Notch signaling exhibits subtype-specific metabolic regulation in GC. The USP24/PLK1/Notch1 axis (144) stabilizes PLK1 via USP24-mediated deubiquitination, phosphorylating Notch1 to activate HES1/HEY1-driven glycolysis and cell cycle progression. Clinically, USP24 overexpression correlates with 5-FU resistance and reduced survival. In contrast, the FPR3/NFATc1/NOTCH3 axis (145) exerts tumor-suppressive effects in diffuse-type GC by modulating calcium flux: FPR3 activation restricts NFATc1 nuclear translocation, downregulating NOTCH3-AKT/mTOR signaling and suppressing stemness. Therapeutic exploitation includes USP24 inhibitors combined with PI3K inhibitors in intestinal-type GC, versus FPR3 agonists in diffuse-type tumors, guided by single-cell EMT and immune subtype profiling.

The Hippo/YAP pathway emerges as a master metabolic regulator. Ginkgolide (146), a natural compound, activates LATS1/2 to phosphorylate YAP and degrade TAZ, suppressing YAP/TAZ-mediated transcription of GLUT1 and LDHA. In PDX models, Gyp reduces tumor volume by 76% and FDG-PET SUVmax by 82%, with nuclear YAP phosphorylation serving as a predictive biomarker for therapeutic response. Clinically relevant combinatorial regimens pairing YAP/TAZ inhibitors with mTOR blockers enhance efficacy while mitigating compensatory signaling.

Metabolic vulnerabilities in GC converge with chemoresistance through dynamic adaptations in glycolysis, redox balancing, and mitochondrial-nuclear crosstalk. Glycolytic drivers such as lncRNA SNHG16 and HKDC1 rewire energy metabolism to sustain drug efflux and DNA repair, while stemness reprogramming via MCM10 and RORα loss couples metabolic plasticity with apoptotic evasion. Therapeutic strategies targeting these axes exploit multimodal approaches—nanoparticle-mediated enzyme degradation, metabolic-immune niche disruption, and pan-pathway inhibition—to induce synthetic lethality. Synergistic regimens combining OXPHOS/glycolysis blockade or HDAC inhibitors with metabolic modulators demonstrate rapid efficacy, validated by advanced imaging biomarkers. The molecular hierarchies of these resistance mechanisms and corresponding therapeutic interventions are systematically mapped in Figure 5 and Table 4 , providing a blueprint for overcoming GC’s adaptive metabolic resilience through precision targeting.

Emerging therapeutic paradigms in GC are uncovering novel regulatory layers spanning microbial ecosystems, nuclear metabolic compartmentalization, and stress-responsive redox circuits. Microbiota-metabolite interactions reshape chemosensitivity through metabolite-driven transcriptional repression of glycolytic effectors, while transcriptional-epigenetic networks enforce metabolic addiction via chromatin remodeling and Ca²⁺ signaling activation. Spatial reorganization of nuclear enzyme complexes under therapy stress creates drug-resistant niches, and redox adapters modulate nutrient deprivation responses through autophagy-microtubule dynamics. These mechanisms are exploitable via engineered probiotics, lactylation-targeted epigenetic editing, nuclear HSP90 inhibitors, and redox disruptors. The molecular architecture and therapeutic vulnerabilities of these emerging targets, alongside their clinical translation potential, are comprehensively detailed in Table 5 , providing a roadmap for next-generation interventions against GC’s adaptive survival programs.

The systematic dissection of metabolic reprogramming in GC has unveiled intricate molecular networks that sustain tumor progression and therapeutic resistance. Mechanistic advances delineating the crosstalk between oncogenic signaling, epigenetic remodeling, and microenvironmental adaptation have positioned metabolic plasticity as a linchpin of malignant survival (159–163). Key regulatory nodes within glycolytic flux, redox balancing, and stress response pathways—including HIF-α/AMPK/mTOR signaling, RNA-modified enzyme complexes, and microbiota-metabolite interactions—now offer actionable targets for therapeutic exploitation (121, 145, 164–167). Emerging strategies extend beyond enzymatic inhibition to encompass microbial modulation and spatial metabolic disruption, reflecting the multidimensional nature of cancer metabolism (168–170). Despite significant advancements in GC research, several critical barriers remain in translating preclinical findings to clinical applications. One major challenge is the inability of current preclinical models to replicate the metabolic heterogeneity observed across different GC molecular subtypes and metastatic microenvironments. GC tumors undergo extensive metabolic reprogramming, yet preclinical models often fail to reflect the complexity of these processes, which vary depending on factors such as genetic mutations and the tumor microenvironment. Additionally, the dynamic interplay between metabolic changes and immune evasion, particularly in therapy-resistant niches enriched with immunosuppressive cells like regulatory T cells (Tregs) and M2 macrophages, is still not fully understood. The lack of standardized metabolic profiling protocols further complicates the comparison of results across studies, limiting the clinical applicability of findings. Addressing these challenges requires the development of standardized techniques for metabolic profiling, which could enable the identification of patients most likely to benefit from targeted therapies.

To overcome these hurdles, a multifaceted approach is needed, involving clinical innovation, technological advancements, and a deeper mechanistic understanding of tumor biology. Clinical trials, particularly those in phase II/III, should consider integrating metabolic inhibitors—such as LDHA blockers or SIRT1 activators—with chemotherapy or immunotherapy. These studies could benefit from biomarker-enriched cohorts, utilizing advanced tools like hyperpolarized ¹³C-MRI or ctDNA-based metabolic signatures for real-time monitoring of therapeutic efficacy. Furthermore, emerging technologies such as spatial metabolomics and single-cell flux analysis could provide detailed insights into the metabolic dependencies of GC subtypes, facilitating the identification of subtype-specific vulnerabilities for targeted therapy. AI-driven multi-omics integration could also help predict adaptive resistance mechanisms, such as the reactivation of OXPHOS in response to therapy, offering new avenues for overcoming resistance. Mechanistically, future research should focus on how tumor-derived metabolites—such as lactate and ketone bodies—interact with immune cells to modulate immune responses, particularly in relation to T cell exhaustion and immune evasion. For example, succinate accumulation in SDH-deficient GC may activate SUCNR1 on dendritic cells, suppressing antitumor immunity. Although still in preclinical development, SUCNR1 antagonists are being explored as potential therapeutic agents to reverse this immune suppression. Additionally, strategies such as nanoparticle-mediated delivery of HK2 inhibitors to TAMs could help reverse their immunosuppressive polarization, enhancing systemic immunity while targeting the tumor microenvironment. These integrated approaches, combining metabolic reprogramming with immune modulation, offer promising directions for developing more effective and personalized therapies for GC in the future.

PKM2, a glycolytic enzyme, has been found to be highly expressed in many cancers, including GC. In the context of exosomes, PKM2 - loaded exosomes released by cancer cells can be taken up by neighboring cells, such as macrophages in the tumor microenvironment. Macrophages, which are integral components of the immune system, can be polarized into different phenotypes in the tumor setting. Tumor - associated macrophages (TAMs) often exhibit an M2-like phenotype that promotes tumor growth, invasion, and metastasis. Emerging data suggest that exosomal PKM2 plays a role in activating lipid synthesis pathways in macrophages. Once macrophages internalize exosomal PKM2, it can modulate intracellular signaling pathways. PKM2 may act as a protein kinase in addition to its enzymatic function. It can phosphorylate key regulators involved in lipid metabolism. For example, it might phosphorylate transcription factors that are crucial for the activation of genes encoding enzymes in the lipid synthesis pathway, such as acetyl - CoA carboxylase (ACC) and fatty acid synthase (FAS). Activation of these enzymes leads to increased de novo lipid synthesis in macrophages (171).

Increased lipid synthesis in macrophages has several implications for GC progression. Lipids can serve as building blocks for cell membranes, which is important for the rapid proliferation of cancer cells. Macrophages with enhanced lipid synthesis can secrete lipid - rich vesicles or directly transfer lipids to cancer cells, providing them with the necessary resources for growth and division. Moreover, lipids can also function as signaling molecules. For instance, certain lipid species can activate signaling pathways in cancer cells that promote their migration and invasion, processes that are critical for GC metastasis. Furthermore, the activation of lipid synthesis pathways in macrophages by exosomal PKM2 may also contribute to the immunosuppressive tumor microenvironment. M2 - like macrophages are known to secrete cytokines and chemokines that can inhibit the function of immune effector cells, such as T - lymphocytes. By enhancing lipid synthesis, exosomal PKM2 may further skew macrophages towards an immunosuppressive phenotype, thus allowing cancer cells to evade immune surveillance (172). Looking ahead, emerging technologies offer promising avenues to target the exosomal PKM2-lipid synthesis axis in gastric cancer. AI-based metabolic imaging, for instance, could enable non-invasive visualization of lipid metabolic reprogramming in tumor-associated macrophages (TAMs) within the gastric cancer microenvironment. By integrating high-resolution imaging data with machine learning algorithms, this approach might precisely map spatiotemporal changes in lipid synthesis triggered by exosomal PKM2, facilitating early detection of microenvironmental dysfunction and personalized treatment monitoring. Meanwhile, CRISPR-based metabolic reprogramming presents another frontier. Engineered CRISPR systems could be tailored to disrupt PKM2 expression or its downstream signaling in macrophages, directly inhibiting exosome-induced lipid synthesis pathways. Such strategies might selectively reverse the pro-tumorigenic phenotype of TAMs without disrupting systemic lipid metabolism, offering a novel precision therapy to normalize the tumor microenvironment and enhance responsiveness to existing treatments.

Overall, glycolytic reprogramming is a hallmark of GC progression, driving therapy resistance and metabolic adaptability. Key regulatory nodes in glycolysis—from HK-mediated glucose trapping to lactate-fueled microenvironment remodeling—offer promising therapeutic targets. While preclinical advances in targeting glycolytic enzymes and signaling hubs demonstrate efficacy, clinical translation requires overcoming tumor heterogeneity and adaptive resistance mechanisms. Future progress hinges on integrating glycolytic biomarkers, advanced metabolic imaging, and innovative delivery systems to enable precision targeting of this central metabolic vulnerability, ultimately reshaping therapeutic paradigms for GC.