Lactate-lactylation in tumor angiogenesis and progression: mechanisms, biomarker potential, and therapeutic implications

Abstract

Tumor growth and metastasis are critically dependent on the tumor vasculature, which provides essential nutrients and oxygen. Metabolic reprogramming—a hallmark of cancer—drives aggressive progression in solid tumors and is characterized by excessive lactate production. Beyond its role as a glycolytic byproduct, lactate functions as a versatile metabolic substrate and signaling molecule that orchestrates tumor angiogenesis and sustains the immunosuppressive vascular niche. The recent discovery of lysine lactylation has unveiled an epigenetic mechanism through which lactate directly regulates gene expression, thereby bridging metabolic activity with pro-tumorigenic transcriptional programs. Targeting key nodes in lactate metabolism—including biosynthetic enzymes, membrane transporters, and lactylation-modifying machinery—holds substantial promise for biomarker development and therapeutic intervention. Notably, conventional anti-angiogenic therapies often face limitations such as transient efficacy, adaptive resistance, and exacerbation of immunosuppression. In contrast, disrupting lactate production, transport, and lactylation offers a multimodal strategy to reprogram tumor metabolism, normalize aberrant vasculature, and reactivate antitumor immunity. Therefore, We propose that concurrently targeting lactate production, transport, and lactylation constitutes a multimodal therapeutic strategy. By reprogramming tumor metabolism, normalizing the vasculature, and reinvigorating antitumor immunity, this integrated approach may surmount the limitations of current anti-angiogenic therapies and yield more durable clinical outcomes.

Graphical Abstract

Introduction

Once solid tumors exceed a certain volume (approximately 1–2 mm³), they become reliant on a newly formed vascular network to supply oxygen and nutrients and to efficiently remove metabolic waste; otherwise, they enter a state of growth arrest or undergo necrosis [1]. The tumor vasculature not only provides essential support for tumor cell survival and proliferation but also serves as a critical conduit for invasion and distant metastasis.

During tumor progression, stress signals such as hypoxia and nutrient deprivation drive the activation of a “pro-angiogenic switch.” In this process, vascular endothelial growth factor (VEGF) binds to its receptor (VEGFR) on endothelial cells (ECs) [2] subsequently activating downstream signaling pathways such as PI3K/AKT and Ras/MAPK, which promote EC survival and proliferation. Concurrently, the expression of matrix metalloproteinases (MMPs, e.g., MMP-2, MMP-9) is upregulated, degrading the vascular basement membrane and guiding EC migration and sprouting toward hypoxic regions to form new “vascular buds.” Other critical pathways, including Notch/DLL4, Ang-1/Tie-2, and Ephrin/EphR, also coordinately regulate the aberrant construction of the tumor vascular system [3].

Compared to normal mature vasculature, tumor blood vessels often exhibit a disorganized architecture, lacking clear stratification and a well-structured network. These vessels are characterized by abnormal EC proliferation and adhesion, incomplete extracellular matrix coverage, and excessive, chaotic branching. These structural defects lead to increased vascular permeability and inefficient perfusion, collectively fostering a unique tumor microenvironment (TME) marked by hypoxia, low pH, and high lactate levels [4]. This dysfunctional vascular network not only fails to support effective immune cell infiltration but also exacerbates tumor immune evasion and therapy resistance [5].

Metabolic reprogramming is an adaptive change in tumor cells to meet the high energy and biosynthetic demands of rapid proliferation [6] and serves as a key driver of tumor angiogenesis. Even under conditions of sufficient oxygen supply, tumor cells preferentially metabolize glucose via the glycolytic pathway, generating large amounts of lactate—a phenomenon known as “aerobic glycolysis” or the “Warburg effect” [7].

Historically, lactate has been regarded as a terminal “waste” product of glycolysis, with its primary role believed to be acidifying the tumor microenvironment. However, recent studies have progressively revealed that lactate is far more than a metabolic end product. It functions as a crucial energy carrier and signaling molecule, deeply involved in various malignant processes such as tumor growth, metastasis, immune regulation, and therapy resistance [8]. For instance, studies have shown that after injecting mice bearing xenograft tumors with ¹³C-labeled glucose and lactate, the abundance of TCA cycle metabolites derived from lactate can be twice that of those derived from glucose [9]. This finding significantly challenges the traditional view of lactate as merely an anaerobic metabolic byproduct. Notably, in human melanoma xenograft models, highly metastatic tumor cells exhibit significantly increased lactate uptake compared to their low-metastatic counterparts, and their TCA cycle metabolic profiling reveals more active lactate utilization [10], indicating a close association between lactate metabolism and tumor metastatic capacity. Indeed, within the tumor microenvironment, elevated lactate levels can act upon various stromal cells to promote angiogenesis. For example, tumor-associated endothelial cells (TECs) can remodel their own metabolism under lactate stimulation [11], providing energetic support for angiogenesis and directly activating pro-angiogenic signaling pathways to promote new blood vessel formation [12]. Concurrently, lactate can enhance the recruitment and function of immunosuppressive cells. For instance, it promotes the polarization of tumor-associated macrophages (TAMs) toward the M2 phenotype, recruits myeloid-derived suppressor cells (MDSCs), and impairs the cytotoxic function of T cells [13]. These alterations further lead to increased release of pro-angiogenic factors, accelerate extracellular matrix degradation, and affect the expression of adhesion molecules on endothelial cells. Consequently, they hinder the migration of CD8⁺ T cells along the vessel walls, ultimately fostering the formation of immature, immune-privileged, and dysfunctional tumor vasculature. Notably, the recent discovery of lactylation—a novel post-translational modification—provides a new epigenetic perspective on lactate-mediated regulation of tumor angiogenesis. Lactylation refers to the covalent binding of lactate to lysine residues on proteins, thereby modulating protein function and gene expression [14]. This modification can directly influence chromatin structure and accessibility. Studies have confirmed its significant role in tumor angiogenesis [15] and its ability to drive the formation of an immunosuppressive microenvironment [16], further promoting tumor progression.

Notably, lactate levels and the extent of protein lactylation within the tumor microenvironment are emerging as novel biomarkers with significant translational potential. On one hand, circulating lactate levels and the expression of lactate metabolism-related genes have been evaluated as prognostic indicators or predictors of therapeutic response in various cancers [17–19].

On the other hand, lactylation—as a more direct functional biomarker—is increasingly recognized for its value. Studies have found that the lactylation levels of specific proteins (e.g., histones or metabolic enzymes) in tumor or endothelial cells correlate closely with tumor angiogenic activity and metastatic potential [19]. Currently, inhibitors targeting lactate metabolism have entered clinical trials, demonstrating promising prospects. In the future, targeting lactate metabolism and specific lactylation sites, in combination with anti-angiogenic therapy and immunotherapy, represents a potential novel strategy for cancer treatment. (Graphical Abstract).

Lactate homeostasis in the tumor microenvironment

The Warburg effect and lactate accumulation

In 1926, Otto Warburg’s seminal discovery revealed that cancer cells preferentially generate lactate via the glycolytic pathway, even under conditions of adequate oxygen supply, rather than utilizing the more efficient mitochondrial oxidative phosphorylation (OXPHOS) for ATP production [20]. This finding challenged the prevailing biological understanding of the time. In normal cells, the presence of oxygen typically inhibits the glycolytic rate through a process known as the Pasteur effect. The glycolytic end-product pyruvate enters the mitochondria, where it is converted to acetyl-CoA by the pyruvate dehydrogenase complex (PDH) and subsequently oxidized via the tricarboxylic acid (TCA) cycle and OXPHOS to carbon dioxide and water, yielding a substantial amount of ATP as the primary energy source. Oxygen acts as a negative feedback regulator, inhibiting phosphofructokinase-1 (PFK-1), a rate-limiting enzyme of glycolysis, thereby reducing its flux [21].

In contrast, cancer cells exhibit profound metabolic reprogramming. PDH activity is frequently suppressed or downregulated. Meanwhile, transcription factors such as c-Myc and hypoxia-inducible factor-1α (HIF-1α) mediate the upregulation of key glycolytic enzymes, including pyruvate dehydrogenase kinase-1 (PDK1), hexokinase 2, phosphofructokinase-2/fructose-2,6-bisphosphatase [22], and others, steering metabolism toward glycolysis and lactate production. Although the yield of ATP per glucose molecule is significantly lower from aerobic glycolysis (2 ATP) compared to OXPHOS (~ 32 ATP), the glycolytic flux in cancer cells is vastly accelerated to compensate. Glucose is rapidly taken up by cancer cells and, through a series of upregulated enzymatic reactions, converted to pyruvate. Subsequently, pyruvate is predominantly reduced to lactate by lactate dehydrogenase (LDH), particularly the LDH-A isoform [23], with the concurrent oxidation of NADH to NAD⁺ [24, 25]. This process not only provides a rapid pathway for ATP generation but also sustains continuous glycolysis through NAD⁺ regeneration. This metabolic adaptation supports the bioenergetic and biosynthetic demands of rapidly proliferating cells, while helping to maintain redox homeostasis, especially under hypoxic stress or during high proliferative states [26–28].

Lactate transport and the lactate shuttle

Lactate transport and the lactate shuttle represent a core mechanism for coordinating cellular metabolism and regulating the tumor microenvironment. This mechanism is primarily mediated by the monocarboxylate transporter (MCT) family, with SLC16A1 (MCT1) and SLC16A3 (MCT4) playing the most prominent roles. Together, they constitute the key molecules governing the transmembrane movement of lactate [29, 30]. These transporters facilitate the bidirectional exchange of lactate across the plasma membrane via the symport of lactate and protons (H⁺) [31]. Notably, the MCT family is functionally distinct from another class of sodium-dependent lactate cotransporters, such as SLC5A12 (SMCT2) [32]. These two transporter types differ fundamentally in their driving ions, transport directionality, and physiological functions.

MCT4 is a hypoxia-inducible transporter that is highly expressed in cells with active glycolysis, such as tumor cells and cancer-associated fibroblasts, and is primarily responsible for the rapid efflux of lactate from these cells [33, 34]. Conversely, MCT1, a constitutively expressed isoform regulated by transcription factors including c-Myc, is predominantly expressed in cells with active oxidative metabolism, tumor-associated endothelial cells [35], and certain immune cell populations [36, 37]. Its main function is to import extracellular lactate for further metabolic utilization. This coordinated transport mechanism not only helps maintain intracellular pH homeostasis but also contributes to the acidification of the extracellular microenvironment, indirectly affecting cancer cell survival and proliferation. It is important to note that cancer cells with extremely high glycolytic activity often exhibit a limited capacity for lactate uptake and/or oxidation under normoxic conditions [38, 39]. This phenomenon can be attributed to several factors. Firstly, these cells possess inherently low oxidative metabolic capacity, coupled with sustained high glycolytic flux, resulting in continuous, high-level intracellular lactate production that establishes a strong, outwardly-directed gradient for both lactate and protons (H⁺). Secondly, under these conditions, the proton-coupled symport function of MCTs is driven by the high intracellular concentrations, resulting in a net efflux of lactate. Beyond its role as a metabolic intermediate, lactate functions as a signaling molecule. It can act in an autocrine or paracrine manner by binding to the G protein-coupled receptor GPR81 (HCAR1) [40], thereby regulating various biological processes such as cellular metabolism, survival signaling, angiogenesis, and immune responses [41, 42].

Based on the aforementioned transport mechanisms, the “lactate shuttle” theory further elucidates the dynamic transmission network of lactate as both an energy carrier and signaling molecule between cells, tissues, and even organs. This concept was initially proposed by Brooks et al. in 1985 [43], emphasizing lactate as a crucial metabolic intermediary linking anaerobic glycolysis and aerobic respiration [44]. Within the tumor microenvironment, the lactate shuttle is markedly activated and undergoes “pathological remodeling,” primarily manifesting in four distinct modes [45]: First, metabolic symbiosis between cancer cells. Tumor cells located in hypoxic zones undergo intensive glycolysis via the Warburg effect, generating substantial lactate which is extruded into the extracellular space via MCT4. In contrast, adjacent cancer cells in oxygenated regions import this lactate through MCT1, converting it back to pyruvate via LDHB for subsequent oxidation in the tricarboxylic acid (TCA) cycle to produce energy. This creates metabolic complementarity within the tumor, supporting growth and metastasis in a heterogeneous microenvironment. This model has been morphologically validated in studies of pancreatic neuroendocrine tumors, among others [46]. Second, the reverse Warburg effect between cancer cells and stromal cells. In this mode, cancer-associated fibroblasts (CAFs) undergo a glycolytic phenotypic switch [47]. Through HIF-1α-dependent epigenetic reprogramming [48] and autophagy-related pathways (e.g., PDGFR-β/Cav-1) [49, 50], CAFs enhance lactate production and efflux. The secreted lactate then serves as a metabolic fuel for neighboring cancer cells and drives tumor progression via signaling axes such as TGF-β1/p38/MMP [51]. Third, the shuttle between the extracellular matrix and endothelial cells. Lactate can be taken up by endothelial cells via surface MCT1 or sensed via GPR81 [40]. This uptake activates pathways like HIF-1α and NF-κB, promoting the expression of angiogenic factors such as VEGF, thereby inducing tumor angiogenesis [52]. Fourth, the shuttle between the extracellular matrix and immune cells. Lactate derived from tumor cells and CAFs accumulates in the microenvironment [53]. Immune cells perceive this lactate signal through MCTs or G protein-coupled receptors (e.g., GPR81, GPR132) [54]. This lactate sensing subsequently drives macrophages toward M2 polarization, suppresses cytotoxic T cell function, enhances regulatory T cell activity, and recruits myeloid-derived suppressor cells, collectively shaping an immunosuppressive microenvironment [41]. Therefore, lactate transport and shuttling represent not merely a key metabolic adaptation but also a core mechanism enabling intercellular metabolic coupling, microenvironmental acidification, aberrant angiogenesis, and immune evasion in pathological states such as cancer (Fig. 1).

Fig. 1.

Lactate production, shuttling, and isomers in tumors microenvironment. Module A: Warburg Effect and Lactate Production. Even under oxygen-sufficient conditions, cancer cells preferentially undergo aerobic glycolysis (the Warburg effect). Glucose is metabolized via glycolysis to pyruvate. Due to inhibition of the PDH complex, pyruvate is largely prevented from entering mitochondrial oxidative phosphorylation. Instead, it is extensively reduced to L-lactate by LDH-A catalysis. This process is coupled with NAD⁺ regeneration, sustaining the high glycolytic flux and forming the metabolic basis of the tumor’s “lactate storm.” Module B: Lactate Transport and Shuttling Networks. Lactate orchestrates intricate intercellular communication via monocarboxylate transporters (MCT1/MCT4), establishing four key shuttling modes: (1) Metabolic symbiosis among cancer cells – Hypoxic cancer cells export lactate, which is utilized by cancer cells in oxygen-rich regions. (2) Reverse Warburg effect – Activated cancer-associated fibroblasts (CAFs) fuel cancer cells by providing lactate. (3) Endothelial cell shuttling – Lactate uptake or signaling via MCTs or GPCRs in endothelial cells activates the HIF-1α/NF-κB-VEGF axis, driving angiogenesis. (4) Immune cell shuttling – Immune cells perceive lactate signals via MCTs or GPCRs (e.g., GPR81, GPR132). Lactate inhibits T-cell function, drives macrophage M2 polarization, and thereby shapes the immunosuppressive microenvironment. Module C: Unique Sources and Functions of D-Lactate. This section comparatively illustrates the structural differences and unique biological properties of D-lactate versus L-lactate. Primarily derived from gut microbial metabolism and the intracellular methylglyoxal pathway, D-lactate exists at much lower concentrations than L-lactate but exhibits distinct functions: it can induce ferroptosis in cancer cells, promote repolarization of macrophages towards the M1 phenotype, and serve as a lactyl donor mediating novel protein D-lactylation modifications, such as modifying RelA (p65) to inhibit NF-κB-mediated inflammatory signaling

D-lactate: an overview and its role in cancer

Initially, lactylation was identified as a modification derived from the L-enantiomer of lactate (Kla). However, the D-enantiomer form of lysine lactylation (Kdla) also exists in cells, with both modifications originating from glycolytic metabolites. In mammalian cells, L-lactate is the primary end product of glycolysis, generated from pyruvate reduction catalyzed predominantly by lactate dehydrogenase A (LDH-A). Its concentration under physiological and pathological conditions ranges from sub-millimolar to tens of millimolar, far exceeding that of D-lactate (< 20 µM) [55]. In contrast, D-lactate is present at very low abundance in the human body. It is primarily generated via two pathways: first, through the methylglyoxal pathway (the glyoxalase system) under conditions of aberrant metabolism in eukaryotic cells [56]; and second, as a metabolic byproduct of gut microbiota (e.g., Lactobacillus and Bifidobacterium), which can be absorbed into the systemic circulation [57]. Although structurally similar, L- and D-lactate exhibit significant differences in their metabolic pathways and physiological functions. Regarding the regulation of cancer cell fate, D-lactate has been shown to induce ferroptosis in esophageal squamous cell carcinoma (ESCC). The underlying mechanism involves D-lactate accumulation triggering oxidative stress, glutathione depletion, and iron overload. Conversely, tumor cells can activate the “CDK7-YAP-LDHD” axis to eliminate D-lactate and sustain survival, highlighting a direct link between D-lactate levels and cancer cell fate [58]. In tumor immunomodulation, D-lactate can inhibit the PI3K/AKT signaling pathway in hepatocellular carcinoma-associated macrophages, driving their repolarization toward the anti-tumor M1 phenotype. This shift is characterized by the downregulation of M2 markers (e.g., ARG1, Fizz1, and IL-10) and the upregulation of M1 markers (e.g., TNF-α, NOS2, and IL-12), thereby reshaping the immunosuppressive tumor microenvironment [59, 60]. Furthermore, D-lactate exerts dual effects on tumor cell energy metabolism. Cai et al. reported that D-lactate selectively induces mitochondrial oxidative phosphorylation and reversibly inhibits aerobic glycolysis in cancer cells in an ATP-dependent manner, while simultaneously promoting the proliferation and effector functions of primary T cells [61]. Conversely, other studies suggest that, similar to L-lactate, D-lactate may enhance DNA repair capacity in cervical cancer cells by inhibiting histone deacetylases and activating the HCAR1 receptor, potentially influencing chemotherapy resistance [62]. Although L-lactylation currently dominates lactylation research, lactoylglutathione (LGSH), generated via the glyoxalase pathway, can also serve as a direct lactyl donor, transferring a D-lactyl group to lysine residues of target proteins to form D-lactylation (Kdla) [63]. This modification possesses unique biological functions. For instance, D-lactate-mediated lactylation at the K310 site of the transcription factor RelA (p65) can suppress NF-κB transcriptional activity and reduce the expression of downstream inflammatory cytokines such as IL-6 and TNF-α [64]. This finding provides a novel perspective for understanding the role of D-lactate in cancer.

Lactate-driven metabolic reprogramming in TECs promotes tumor angiogenesis

Lactate plays a dual and central role in the tumor microenvironment, both driving metabolic reprogramming in endothelial cells and promoting tumor angiogenesis, thereby forming a multi-layered, self-amplifying network (Fig. 2). As key executors of angiogenesis, tumor endothelial cells (TECs) exhibit high metabolic plasticity, transitioning from a quiescent state dependent on mitochondrial oxidative phosphorylation to a glycolytic-dominant metabolic phenotype, supplemented by enhanced glutamine metabolism and fatty acid oxidation, to meet the energetic and biosynthetic demands of rapid proliferation and migration [11, 65]. This “glucose-addicted” state results in over 85% of ATP being generated via glycolysis, accompanied by substantial lactate production [66] However, lactate is not merely a metabolic end-product but also a critical signaling molecule. By inhibiting prolyl hydroxylase activity, lactate stabilizes hypoxia-inducible factor-1α (HIF-1α), which in turn transcriptionally upregulates the expression of glucose transporter 1 and various glycolytic enzymes. This establishes a “lactate-HIF-1α-glycolysis” positive feedback loop, locking endothelial cells into a pro-angiogenic phenotype [66]. Within this process, overexpression of PFKFB3—a key regulatory node in glycolysis—is crucial for the migration and guidance functions of tip cells [67] and helps maintain the angiogenic cell fate by inhibiting the Notch signaling pathway [65, 68]. This metabolic shift is further reinforced at the transcriptional level, where inhibitory transcription factors such as KLF2/FOXO1 are suppressed, allowing pro-glycolytic transcription factors like MYC to become activated [69, 70].

Fig. 2.

The network of lactate, tumor endothelial cell metabolic reprogramming, and angiogenesis. A. Lactate-induced endothelial metabolic reprogramming: The diagram illustrates how lactate stabilizes HIF-1α by inhibiting PHD2, driving endothelial cells to shift from oxidative phosphorylation (OXPHOS) to a glycolysis-dominant metabolism, forming a “lactate-HIF-1α-glycolysis” positive feedback loop. The upregulation of the glycolytic key enzyme PFKFB3 promotes tip cell migration and inhibits Notch signaling. Transcription factors KLF2/FOXO1 are downregulated, while MYC is activated, locking endothelial cells into a pro-angiogenic phenotype. B. Lactate as a metabolic hub: Lactate serves as a central metabolic coordinator, integrating multiple metabolic networks: ① Stabilizing HIF-2α to promote glutamine metabolism, forming a “lactate-glutamine” cycle and regenerating NADPH. ② Providing acetyl-CoA to enhance fatty acid synthesis and regulating lipolysis via ApoC2-K70 lactylation. ③ Metabolites synergize with glutamine to activate mTORC1, promoting the synthesis of proteins such as VEGFR2. C. Paracrine regulation of tumor-derived lactate: Three representative tumor pathways (gastric cancer HOXA9/c-MYC, bladder cancer FGF6/PI3K/Akt, and colorectal cancer DKK2/mTOR) all upregulate glycolysis in tumor cells to increase lactate production. Lactate acts on endothelial cells in a paracrine manner to promote migration, tube formation, and proliferation. Each pathway is validated through gene knockdown, inhibitor treatment, and lactate rescue experiments, confirming lactate as the terminal effector molecule promoting angiogenesis. D. Dual regulation of vascular function by lactate: Lactate exhibits both pro-angiogenic and pro-metastatic functions: - Lactate activates TGF-β signaling to induce endothelial-to-mesenchymal transition (EndMT), downregulates tight junction proteins (ZO-1, Occludin), and disrupts vascular barrier integrity. - Tumor-derived exosomes (e.g., miR-221-3p) can enhance glycolysis and lactate production by suppressing LIFR, further amplifying barrier damage and promoting metastasis

While this metabolic transition is often described as a general paradigm, significant heterogeneity likely exists in the metabolic plasticity and responsiveness to lactate among endothelial cells from different organs, at various differentiation stages, or of distinct vascular types (e.g., arterial, venous, lymphatic). Notably, tumor cell-derived lactate functions in a paracrine manner to modulate endothelial cell function, driving tumor angiogenesis and metastasis. In gastric cancer, the highly expressed HOXA9 activates the transcription factor c-MYC, upregulating key glycolytic enzymes (e.g., HK2, GLUT1, LDHA) and promoting lactate production. This lactate-rich environment significantly stimulates lymphangiogenesis and endothelial cell migration. HOXA9 knockdown reduces lymph node metastasis, an effect that can be reversed by exogenous lactate supplementation [71]. In bladder cancer, fibroblast growth factor 6 promotes tumor cell glycolysis and lactate secretion by regulating the PI3K/Akt and MAPK signaling pathways. The resulting lactate effectively enhances the survival and migration of human umbilical vein endothelial cells (HUVECs), thereby driving angiogenesis [72]. In colorectal cancer, the DKK2 protein accelerates lactate secretion via activation of the PI3K/Akt/mTOR pathway. Lactate, acting as its downstream effector, stimulates endothelial cells to form tubular structures and enhances their motility. Inhibition of lactate efflux using monocarboxylate transporter (MCT) inhibitors significantly suppresses this pro-angiogenic process [73]. These findings collectively establish lactate as a central hub within the signaling network that promotes angiogenesis in the tumor endothelium. Lactate exerts bidirectional regulatory effects on endothelial cells: while promoting angiogenesis, it also impairs normal vascular structure and function. Lactate supplies energy and biosynthetic precursors to endothelial cells and, through the activation of specific signaling pathways (e.g., PI3K/Akt), directly drives their proliferation, migration, and tube formation. Moreover, lactate can induce endothelial-to-mesenchymal transition (EndoMT) by activating pathways such as TGF-β. This is characterized by decreased expression of endothelial markers, increased expression of mesenchymal markers, and downregulation of tight junction proteins (e.g., ZO-1, Occludin) [74]. Similarly, breast cancer-derived exosomal miR-221-3p promotes glycolysis and lactate production in endothelial cells by targeting the LIFR receptor, subsequently inhibiting tight junction protein expression [69, 70] through a mechanism also associated with TGF-β pathway activation [75]. The disruption of vascular barrier integrity significantly increases vascular permeability, creating favorable conditions for tumor cells to enter the circulation and establish distant metastases.

Furthermore, lactate’s function extends far beyond being a mere end-product of glycolysis. It acts as a metabolic hub, coupling and coordinating amino acid and lipid metabolism to establish an anabolic network. At the level of amino acid metabolism, lactate stabilizes hypoxia-inducible factor-2α (HIF-2α), which in turn induces c-Myc expression. This leads to the upregulation of the glutamine transporter ASCT2 and the key metabolic enzyme glutaminase 1 (GLS1), thereby enhancing glutaminolysis [76]. The α-ketoglutarate (α-KG) generated from glutamine metabolism replenishes the tricarboxylic acid (TCA) cycle. The resulting malate is then converted by malic enzyme (ME1) into NADPH and pyruvate, with the latter potentially being reconverted to lactate, forming a metabolic cycle. Additionally, pyruvate derived from lactate oxidation can contribute to mTORC1 activation through TCA cycle intermediates such as aspartate, synergizing with glutamine [77, 78]. This process drives the synthesis of pro-angiogenic proteins, including VEGFR2 [79].

In terms of lipid metabolism, lactate provides the precursor acetyl-CoA for fatty acid synthesis and upregulates the activity of acetyl-CoA carboxylase (ACC) [80–83]. Studies demonstrate that lactate promotes lipid droplet accumulation by upregulating key fatty acid synthesis enzymes, such as ACLY and ACC [82, 84]. It can also drive extracellular lipolysis through non-histone lactylation modifications—for instance, lactylation at the K70 site of apolipoprotein C-II—thereby releasing free fatty acids (FFAs) to fuel fatty acid oxidation (FAO) in TECs [83, 85]. Moreover, high lactate levels compete with fatty acids for entry into the TCA cycle, inhibiting lipolysis and promoting its own oxidation [84]. Collectively, these mechanisms establish a lactate-centered anabolic network, ensuring a sustained supply of energy, reducing equivalents, and macromolecular building blocks for endothelial cells during the process of angiogenesis.

Lactate modulates tumor immunity to shape the angiogenic landscape

Tumor angiogenesis and tumor immunity are interdependent processes; aberrant vasculature nourishes immune suppression, which in turn reinforces angiogenesis [85]. Lactate has emerged as a core immunomodulatory signaling molecule. It acidifies the microenvironment and acts as a “metabolic instruction,” profoundly reshaping the intratumoral immune landscape and, in turn, sculpting an angiogenic niche conducive to tumor progression (Fig. 3).

Fig. 3.

Lactate as a metabolic hub driving the malignant cycle of immunosuppression and angiogenesis. Tumor cells produce large quantities of lactate through enhanced glycolysis (Warburg effect) and the catalytic action of lactate dehydrogenase A (LDHA), which is then exported via monocarboxylate transporter 4 (MCT4). This leads to acidification of the tumor microenvironment (TME) and the accumulation of lactate. The accumulated lactate acts as a key signaling molecule, remodeling the tumor immune microenvironment through multiple pathways. Lactate activates signaling pathways such as mTOR/HIF-1α/STAT3 via G-protein-coupled receptors (e.g., GPR81, GPR132), driving tumor-associated macrophages (TAMs) toward an M2 phenotype and upregulating their expression of programmed death-ligand 1 (PD-L1) and arginase 1 (Arg1). Simultaneously, lactate enhances the stability and suppressive function of regulatory T cells (Tregs) by upregulating forkhead box protein P3 (FOXP3) and increases their PD-1 expression. Lactate also activates myeloid-derived suppressor cells (MDSCs). Together, lactate synergizes with M2 TAMs, MDSCs, and Tregs to suppress the function and infiltration of CD8⁺ T cells and natural killer (NK) cells. High-lactate tumor vascular endothelial cells enter a functionally suppressed state, downregulating T-cell adhesion molecules (e.g., intercellular adhesion molecule-1, ICAM-1; vascular cell adhesion molecule-1, VCAM-1) while upregulating inhibitory molecules (e.g., PD-L1, Fas ligand, FASL) and selective recruitment ligands (e.g., Clever-1). This hinders the extravasation of effector T cells and promotes the recruitment of immunosuppressive cells. Finally, the immunosuppressive cell populations reshaped by lactate (M2 TAMs, Tregs, MDSCs) collectively secrete a large amount of pro-angiogenic factors (e.g., vascular endothelial growth factor, VEGF; fibroblast growth factor-2, FGF-2; matrix metalloproteinase-9, MMP-9; transforming growth factor-β, TGF-β; interleukin-10, IL-10), forming a “cytokine storm” that strongly stimulates endothelial cell proliferation, migration, and lumen formation. This ultimately leads to structurally and functionally aberrant tumor angiogenesis. The newly formed abnormal vessels provide nutrients to the tumor, further supporting tumor growth and high glycolysis, which continuously generate lactate. This establishes a self-perpetuating “lactate-immunosuppression-angiogenesis” positive feedback loop that drives tumor progression

The immunosuppressive phenotype of tumor endothelial cells

As “gatekeepers” for immune cell trafficking, the functional state of endothelial cells directly determines whether immune cells can effectively infiltrate tumor tissue. Normal transendothelial migration of immune cells relies on the precise coordination of adhesion molecules (e.g., intercellular adhesion molecule-1 [ICAM-1], vascular cell adhesion molecule-1 [VCAM-1]) and chemokines [86, 87]. However, within the TME, tumor endothelial cells (TECs) often downregulate these key adhesion molecules and enter a state of “endothelial anergy,” characterized by a diminished response to inflammatory stimuli. This severely impairs the extravasation and recruitment of anti-tumor immune cells, such as effector T cells [88, 89]. Notably, this suppression is selective. TECs conversely facilitate the infiltration of immunosuppressive cells like monocytes, macrophages, and regulatory T cells (Tregs) by expressing specific molecules such as Clever-1, CX3CL1, and PLVAP [90–92]. Furthermore, TECs actively express inhibitory molecules including PD-L1 and indoleamine 2,3-dioxygenase (IDO), as well as apoptosis-inducing ligands like Fas ligand (FASL). They also secrete immunosuppressive factors such as TGF-β and IL-10, directly suppressing or eliminating cytotoxic T cells [93–96]. Beyond limiting immune cell infiltration, TECs actively contribute to establishing an immunosuppressive milieu. For instance, they express apoptosis-inducing ligands like FASL and TRAIL [97, 98] to trigger cytotoxic T cell death, while Tregs expressing anti-apoptotic proteins like cFLIP are spared. Additionally, TECs upregulate ligands like RAE-1ε, which trigger the internalization of the NKG2D receptor in natural killer (NK) cells, thereby inhibiting their anti-tumor activity [99]. The expression of nitric oxide synthase 2 (NOS2) and NOS3 in TECs also promotes the polarization of tumor-associated macrophages toward a protumor “M2-like” phenotype, further reinforcing the local immunosuppressive environment.

Lactate sculpts the immune-vascular niche

The high concentration of lactate produced by glycolytic cells within the TME is a primary driver of microenvironmental acidosis [100]. This acidic environment itself can induce dysfunction in infiltrating effector T cells [101]. More importantly, lactate acts as an instructive signal that sculpts the immunosuppressive niche, synergizing with and reinforcing the endothelial-mediated immunosuppression described above through multiple mechanisms. As a key immunometabolite, lactate regulates the differentiation, polarization, and function of various immune cells within the TME [5], skewing the immune landscape toward a pro-angiogenic state. Macrophages: Lactate potently promotes the polarization of tumor-associated macrophages (TAMs) toward a pro-angiogenic M2-like phenotype [102, 103]. This involves diverse signaling pathways. For example, lactate inhibits TFEB-mediated expression of ATP6V0d2 via mTORC1, reducing the degradation of HIF-2α and thereby enhancing the expression of pro-angiogenic genes in macrophages [104]. In pituitary adenomas, lactate promotes M2 polarization of TAMs via the mTORC2-Akt axis, which in turn exacerbates tumor invasion through CCL17-CCR4-mTORC1 signaling [105]. Lactate can also phosphorylate and activate the ERK/STAT3 pathway, promoting M2 polarization, upregulating Arg1 and Mrc1 expression, and facilitating breast cancer cell proliferation, migration, and angiogenesis [106, 107]. Lactate further modulates macrophage function via proton-sensing G protein-coupled receptors (GPCRs) [108]. In breast cancer, extracellular lactate activates GPR132 (G2A) and GPR65 (TDAG8), which are highly expressed on macrophages [109]. This triggers the cAMP-ICER pathway, inducing M2 polarization, upregulating chemokines like CCL17, CCL22, and toll-like receptor 1 (TLR1), and correlating with cancer cell adhesion, invasion, and metastasis [110]. In various tumors (including breast, lung, and colorectal cancers) [111, 112], lactate binding to GPR81 on macrophages downregulates the cAMP-PKA pathway, upregulates PD-L1 expression, promotes M2-like polarization, and mediates immunosuppression and angiogenesis [113]. Myeloid-Derived Suppressor Cells (MDSCs): Beyond macrophages, lactate binds to GPR81 on MDSCs [114], activating them via the GPR81/mTOR/HIF-1α/STAT3 pathway [115] promoting their differentiation and enhancing PD-L1 expression [116] NK Cells: Lactate can inhibit NK cell infiltration. Transcriptomic profiling identified upregulation of genes like NR6A1, OSBP2, and UNC119B as mediators of lactate-driven suppression of NK cell recruitment [117] T Cells: The high-lactate, low-pH environment of the TME is closely linked to impaired T cell function and proliferation [118]. Elevated lactate levels inhibit lactate efflux in CD8⁺ T cells, inducing a quiescent state [36], decreasing NAD⁺ levels, and reducing metabolic intermediates required for proliferation, thereby suppressing cytotoxicity [119]. Compared to effector T cells, Tregs more readily access tumor tissue via the abnormal vasculature [120]. Lactate is crucial for Treg proliferation and function. It enhances Treg stability and suppressive function by upregulating FOXP3 expression, granting Tregs a survival advantage in the glucose-depleted, lactate-rich TME [121]. Beyond metabolic adaptation, lactate can also upregulate PD-1 levels on Tregs, enhancing their immunosuppressive activity [122].

Pro-angiogenic effects of immunosuppressive cells

The immune response within the TME can broadly be categorized as pro-tumor (which is often pro-angiogenic) or anti-tumor (which is often anti-angiogenic). The pro-tumor, pro-angiogenic axis is typically mediated by MDSCs [123], M2-like macrophages [124], pro-tumor neutrophils [125], and Tregs [126]. The anti-tumor, anti-angiogenic axis is characterized by the functions of M1-like macrophages [127], natural killer cells [128], and effector T cells [129]. The net outcome of angiogenesis is determined by the balance between these opposing immune forces [130]. M2-like TAMs integrate signals from hypoxia and high lactate, activating MAPK signaling and downstream enzymes like arginase-1 (ARG1) to promote malignant phenotypic transformation [131, 132]. Lactate-polarized macrophages increase VEGF production, forming a positive feedback loop that further stimulates angiogenesis. Specific deletion of the glycolytic regulator Pfkfb3 in endothelial cells reduces their lactate secretion, leading to decreased M2 polarization and impaired pro-angiogenic and pro-regenerative capacity of macrophages [133], underscoring the instructive role of lactate in this process. Furthermore, M2 TAMs secrete MMPs [134], serine proteases, and histones [135], activating endothelial cell signaling pathways (ECS) and contributing to vascular abnormalities, excessive branching, and increased permeability [136]. MDSCs promote angiogenesis by expressing pro-angiogenic factors like FGF-2 and PDGF in a paracrine manner [137, 139] and by highly expressing MMP-9. These factors, along with extracellular matrix components, stimulate endothelial cell proliferation and increase tumor vascular density [138]. Additionally, MDSCs can differentiate into endothelial-like cells and induce tumor cells to form vasculogenic mimicry (VM) structures, thereby promoting tumor blood supply [139] Tregs not only directly secrete pro-angiogenic factors like VEGF but also modulate other cells, such as tumor-associated macrophages and fibroblasts, by secreting cytokines like TGF-β and IL-10 [140]. This induces these cells to produce additional angiogenic factors [141], further promoting tumor vascularization to supply nutrients and oxygen to tumor cells.

Biphasic regulation of angiogenic signaling by lactate

Lactate concentration in tumor tissue is typically 5–20% higher than in normal tissue [142]. Acidification of the local tumor microenvironment (TME) reduces the binding activity of cell integrins to the extracellular matrix (ECM), downregulates E-cadherin expression, and triggers the activation of proteases such as MMP-9, hyaluronidase-2, and cathepsin B produced by cancer cells. These proteases collectively contribute to the promotion of tumor angiogenesis [143–145]. Furthermore, lactate can activate metabolic switching and the angiogenic switch in endothelial cells via paracrine signaling, driving both HIF-1α-dependent and HIF-1α-independent pathways. This leads to the upregulation of downstream pro-angiogenic factor secretion and stimulates tumor angiogenesis [146, 147] to maintain blood perfusion within the hypoxic TME.

Under normoxic conditions, oxygen-dependent prolyl hydroxylase domain proteins (PHDs) hydroxylate proline residues on HIF-1α, targeting it for rapid ubiquitination and proteasomal degradation via the von Hippel-Lindau tumor suppressor protein (pVHL). However, lactate, oxidized to pyruvate, can bind to and inhibit PHD2 activity, thereby stabilizing HIF-1α. This stabilization induces the expression of multiple downstream pro-angiogenic signals, including VEGFA, bFGF, MMPs, CXCL12, and ANGPT2, which act synergistically with lactate-induced VEGF secretion [148]. While the role of HIF-1α in angiogenesis is well-established, HIF-1α knockdown does not appear to suppress blood supply and proliferation in certain tumor types with PHD2 knockdown, suggesting the existence of alternative HIF-1α-independent angiogenic pathways. Research indicates that pyruvate generated from lactate conversion can directly inhibit the binding of PHD2 to N-Myc downstream regulated gene family member 3 (NDRG3), leading to intracellular NDRG3 accumulation. This drives NDRG3 signaling and promotes angiogenesis and tumor proliferation [149]. Studies in lymphoma support this mechanism: the suppression of tumor growth caused by downregulated lactate metabolism in NDRG3-overexpressing lymphoma cells can be blocked. Additionally, in hepatocellular carcinoma (HCC), NDRG3 knockdown inhibits tumor growth and reduces angiogenesis in the tumor stroma. Beyond the NDRG3-Raf/ERK pathway, NF-κB has been identified as a lactate-responsive transcription factor, revealing the existence of HIF-independent angiogenic pathways linked to glucose metabolism [150]. In tumor endothelial cells, lactate-induced PHD2 inhibition also promotes NF-κB activation, which targets IL-8-mediated angiogenesis. This process can be blocked by anti-IL-8 antibodies, ultimately inhibiting tumor proliferation [52]. In breast cancer, lactate binding to GPR81 activates the PI3K/Akt-CREB pathway and increases the secretion of the pro-angiogenic factor amphiregulin (AREG) [151]. Furthermore, chaperone-mediated autophagy (CMA) has been observed to promote angiogenesis by mediating HUVEC tube formation through the regulation of HK2-dependent lactate production [152], indicating that lactate promotes angiogenesis through multiple pathways.

While current evidence largely suggests that tumor endothelial cells (TECs) exhibit adaptive or even proliferative advantages in high-lactate environments, promoting angiogenesis, some studies report that high lactate or acidic conditions can inhibit angiogenesis. This inhibition is primarily manifested as impaired migration, tube formation, and VEGF signaling. Tumor interstitial lactate often reaches 10–40 mM, accompanied by a pH drop to 6.3–7.0 [153]. Research by Seraina Faes et al. demonstrated that VEGF increases AKT and MAPK phosphorylation at pH 7.4, promoting endothelial cell proliferation and survival. However, at pH 6.4, VEGF fails to promote proliferation/survival or activate downstream AKT signaling, and VEGFR2 expression is reduced. Functionally, this results in impaired endothelial cell migration and proliferation [154]. Therefore, we propose that tumor angiogenesis under lactate-acidic conditions in the TME may exhibit a biphasic nature (Fig. 4). This hypothesis finds preliminary support in experiments by A.K. Pedersen et al. In a model simulating the TME (1% hypoxia, low serum, low glucose, low pH, high lactate), endothelial cell signals related to NHE1, AKT, and protein translation were more strongly downregulated compared to a simple hypoxia model, resulting in restricted proliferation and migration—outcomes contradictory to angiogenesis. This suggests that the combined stress of acidity and lactate exerts an inhibitory pressure on endothelial pro-angiogenic functions [155]. Furthermore, Xu et al. reported that lactate activates endothelial cell GPR81, which in turn inhibits the cAMP/RAF1/MEK signaling pathway. This inhibition induces ERK1/2 phosphorylation, increases calpain activity, promotes VE-cadherin endocytosis and cleavage, and weakens the adhesion and polarity required for tip-stalk cell coordination. Consequently, vascular network structures are disrupted in HUVEC tube formation assays, hindering stable neovessel sprouting and maturation [70]. These findings indicate that the biphasic nature of tumor angiogenesis likely involves complex signaling pathways and cell-cell interactions. However, no study has yet clearly defined a specific pH threshold that strictly governs angiogenic function. Under conditions of lactic acidosis, the ability of TECs to proliferate and sustain pro-angiogenic capacity may highly depend on pH buffering compensation mediated by carbonic anhydrase II (CAII). Dorcas A. Annan et al. found that inhibiting CAII significantly reduced TEC survival and proliferation under lactic acidosis. Conversely, if TECs cannot effectively buffer the lactate-acid load, angiogenic inhibitory effects such as low VEGF-axis responsiveness and reduced TEC proliferation become apparent. This phenomenon suggests that CAII-mediated pH compensation is a potential key mechanism for maintaining partial angiogenic capacity in high-lactate environments.

Fig. 4.

Biphasic model of lactate-mediated regulation of tumor angiogenesis. Within the tumor microenvironment defined by lactate concentration and pH, lactate exhibits biphasic effects on angiogenesis. The pro-angiogenic window (upper left, green gradient) is characterized by moderate lactate levels (< 30 mM) and near-physiological pH (> 6.8), which promote vascular growth. In this window, lactate is taken up by endothelial cells and metabolized to pyruvate, which inhibits PHD2 activity. This stabilizes HIF-1α and activates the NDRG3 and NF-κB pathways, collectively upregulating the secretion of pro-angiogenic factors (e.g., VEGFA, bFGF, IL-8, ANGPT2), thereby enhancing endothelial cell proliferation, migration, and tube formation. Conversely, the anti-angiogenic/suppressive window (lower right, red gradient) is characterized by high lactate concentration (> 20 mM) and low pH (< 6.8), which inhibit angiogenesis. Under these conditions, combined lactate and H⁺ overload impairs VEGF signaling (reducing VEGFR2 expression and attenuating AKT/MAPK phosphorylation), while GPR81 activated by lactate triggers cAMP/RAF1/MEK inhibition, calpain activation, and VE-cadherin cleavage, leading to disruption of vascular integrity and suppression of endothelial cell function. Carbonic anhydrase II (CAII) serves as a key pH buffer, enabling endothelial cells to adapt to acidosis; inhibition of CAII exacerbates cell death under low pH conditions. Therapeutic implications: Targeting lactate metabolism (e.g., using MCT inhibitors) may synergize with anti-VEGF therapies in the anti-angiogenic window, but could attenuate the efficacy of anti-VEGF therapy in the pro-angiogenic window, highlighting the context-dependent nature of metabolic intervention strategies

Lactylation modification in tumor angiogenesis and metastasis

Acute lactate accumulation can rapidly “initiate” an angiogenic response through direct signaling pathways, exerting immediate effects. In parallel, the sustained high-lactate environment further “consolidates” and “amplifies” pro-angiogenic gene expression and metabolic states by inducing and stabilizing lactylation modifications. Lysine lactylation (Kla/Klac), the process by which a lactyl group is transferred from lactate to lysine residues on proteins, including histones [156], significantly enhances the malignant phenotype of tumor cells by regulating gene expression, protein function, and signaling pathways(Fig. 5). This provides a potential mechanism underpinning tumor angiogenesis and metastasis. These two mechanisms—direct signaling and indirect, persistent epigenetic regulation—operate within a dynamic and complementary framework, forming a self-sustaining positive feedback loop in tumor angiogenesis.

Fig. 5.

The lactylation code: writers, erasers, readers and biomarker potential. 1. Writers: Enzymes that catalyze lactylation.p300/CBP: Core lactyltransferases. Catalyze histone H3K18la to drive tumorigenesis (e.g., PDAC, NSCLC) and non-histone lactylation (e.g., YTHDC1) to promote proliferation. AARS1/AARS2: Act as lactate sensors/transferases, lactylating proteins like YAP to activate oncogenic pathways in GC and CRC. KAT Family (KAT2A/7/8): Acyltransferases promoting lactylation in various cancers (e.g., KAT8 in CRC, KAT7 in cervical cancer). Pathological Significance: Their overexpression is frequently linked to poor patient prognosis. 2. Erasers: Enzymes that remove lactylation marks. HDAC1-3: Primary nuclear delactylases. HDAC1/2/3 regulate lactylation balance and therapy resistance. Sirtuins 1–3: NAD⁺-dependent delactylases. SIRT1/2/3 modulate key lactylated oncoproteins (e.g., PKM2, CCNE2). Core Mechanism: Writers and erasers dynamically compete for the same lysine sites (e.g., H3K18), determining the final transcriptional output. 3. Readers: Proteins that specifically recognize and interpret lactylation signals. Key Example: DPF2 preferentially binds H3K14la over acetylation, redirecting transcriptional programs in cervical cancer. Mechanistic Insight: Readers enable metabolic rewiring by decoding lactylation as a distinct signal from acetylation. Biomarker Applications: Diagnosis/Prognosis: High writer (e.g., p300) or low eraser activity often correlates with poor overall survival. Therapy Prediction: Specific lactylation levels (e.g., H3K18la) are associated with resistance to targeted therapies (e.g., EGFR-TKIs) and immunotherapy response. Therapeutic Targets: Targeting this regulatory axis represents a novel strategy to disrupt the metabolism-epigenetics interplay in cancer

Hepatocellular carcinoma (HCC), characterized by abundant microvasculature, provides a clear example. Lactylation at the K430 site of the ABCF1 protein (an ATP-binding cassette transporter) enhances its nuclear translocation. Functioning as a transcription factor, lactylated ABCF1 upregulates HIF-1α expression via the KDM3A-H3K9me2-HIF1α axis, promoting angiogenesis. The resulting increase in lactate levels further enhances ABCF1 lactylation, creating a vicious cycle that drives HCC growth, metastasis, and angiogenesis [157]. Lactylation also promotes angiogenesis by regulating other key molecules and pathways. Ubiquitin-specific peptidase 39 (USP39), known for mediating DNA damage response and radio/chemoresistance by stabilizing checkpoint kinase 2 (CHK2) [158], also promotes malignant proliferation and angiogenesis in renal cell carcinoma by regulating SRSF1 and SRPK1 to suppress the alternative splicing of the anti-angiogenic VEGF-A165b isoform [159]. Recent findings in endometrial carcinoma reveal that histone lactylation promotes USP39 transcription, playing a significant role in metastasis. Specifically, H3K18la enrichment at the USP39 promoter upregulates USP39 expression, which deubiquitinates and stabilizes phosphoglycerate kinase 1 (PGK1). This activates the PI3K/AKT/HIF-1α signaling pathway, enhancing glycolysis and promoting tumor cell proliferation and migration [160]. In esophageal squamous cell carcinoma (ESCC), global Kla levels are markedly elevated and correlate with high invasiveness. Hypoxia-induced H3K9la significantly enriches at the promoter of LAMC2, a gene upregulated in collagen-rich ECM, activating the PI3K/Akt pathway and increasing pro-angiogenic factor expression [161].

Lactylation also profoundly, albeit indirectly, affects the vascular niche by promoting VM—a highly aggressive, endothelial cell-independent mechanism where tumor cells themselves form fluid-conducting, channel-like structures to secure blood supply. Key molecular markers of VM include VEGFR2 and VE-cadherin [162, 163]. In glioblastoma (GBM), the functional peptide P4-135aa, encoded by MAPK6P4, phosphorylates KLF15 at S238, increasing its stability and nuclear translocation. KLF15 acts as an upstream transcriptional activator of LDHA. Overexpressed LDHA binds to and lactylates VE-cadherin, elevating its expression and promoting VM development in GBM [164]. Furthermore, in prostate cancer, MCT1-mediated lactate uptake stabilizes the HIF-1α protein under normoxia, with lactylation signals co-localizing with HIF-1α in an MCT1-dependent manner, suggesting lactylation as a potential HIF-1α stabilization mechanism [165]. HIF-1α binds to and upregulates the transcription of KIAA1199 (CEMIP), which degrades hyaluronic acid to generate pro-angiogenic fragments and regulates factors like VEGFA and VE-cadherin, thereby promoting VM [166]. These observations support a hypothesized positive feedback loop: lactate uptake → HIF-1α lactylation → KIAA1199 upregulation → angiogenesis promotion. However, current evidence is largely based on protein co-localization and expression correlation; biochemical confirmation of HIF-1α lactylation itself, identification of specific lactylation sites via co-immunoprecipitation and mass spectrometry, and discovery of the responsible “writer” enzyme are needed in future studies. Remodeling Tumor Cell Invasiveness and Metastasis: Lactylation also influences angiogenesis by enhancing tumor cell invasion and metastatic capacity. In breast cancer, potassium channel KCNK1 activates LDHA, increasing glycolysis and promoting H3K18la. This lactylation induces the expression of proliferation- and metastasis-related genes (ZWINT, ECT2, ANLN, EZR). The increased LDHA expression acts as a malignant positive feedback mechanism, reducing tumor cell rigidity and adhesion, granting greater deformability for breaching barriers and detaching from tissue, ultimately driving breast cancer proliferation, invasion, and metastasis [167].Colon colorectal cancer (CRC), lactylation promotes metastasis through diverse mechanisms. For example, the lncRNA STEAP3-AS1 drives H3K18la, which promotes the enrichment of the transcription factor ERG at the MMP9 promoter. This degrades the extracellular matrix, disrupts tissue barriers, and paves the way for tumor cell invasion and liver metastasis [168]. Additionally, lactylation of the VIRMA protein, a key component of the m6A methyltransferase complex, increases its activity. This leads to aberrant m6A methylation on SP1 mRNA, elevating SP1 protein levels. SP1 drives TGF-β transcription, creating a dual effect: autocrine action on tumor cells enhances their epithelial-mesenchymal transition (EMT) capacity, while paracrine action on the TME specifically targets ITGA11 + myofibroblast subsets. These activated cancer-associated fibroblasts remodel the ECM and deliver pro-metastatic circTAX1BP1 via exosomes, forming a positive feedback loop driving liver metastasis [169] (Table 1).

Table 1.

Summary of the mechanisms of lactylation in tumor angiogenesis and metastasis

Cancer Type	Modified Target / Key Molecule	Modification Type / Site	Primary Mechanisms and Functions	Impact on Angiogenesis/Metastasis	References
Hepatocellular Carcinoma (HCC)	ABCF1	Lactylation at K430	Promotes nuclear translocation → Acts as a transcription factor → Upregulates HIF-1α via the KDM3A-H3K9me2-HIF1α axis → Elevated lactate further enhances ABCF1 lactylation, forming a vicious cycle.	Promotes tumor growth, metastasis, and angiogenesis	[157]
Endometrial Carcinoma (EC)	Histone H3 / USP39	H3K18la enrichment at the USP39 promoter	H3K18la → Enhances USP39 transcription → USP39 deubiquitinates and stabilizes PGK1 → Activates the PI3K/AKT/HIF-1α pathway → Promotes glycolysis.	Promotes tumor cell development, growth, and migration	[160]
Esophageal Squamous Cell Carcinoma (ESCC)	Histone H3	H3K9la enrichment at the LAMC2 promoter	Hypoxia-induced H3K9la → Upregulates LAMC2 expression → Activates the PI3K/Akt pathway → Upregulates pro-angiogenic factors.	Contributes to high invasiveness	[161]
Glioblastoma (GBM)	VE-cadherin	Protein lactylation	MAPK6P4/P4-135aa → Phosphorylates KLF15 → Activates LDHA transcription → LDHA overexpression binds and lactylates VE-cadherin → Increases its expression.	Promotes vasculogenic mimicry (VM)	[164]
Prostate Cancer (PCa)	HIF-1α (Potential)	Hypothetical lactylation	MCT1-mediated lactate uptake → Stabilizes HIF-1α under normoxia → Lactylation signal co-localizes with HIF-1α (MCT1-dependent) → HIF-1α transactivates KIAA1199 (CEMIP) → Degrades HA into pro-angiogenic fragments and regulates VEGFA/VE-cadherin.	Forms a “Lactate uptake → HIF-1α stabilization/lactylation → Pro-VM” positive feedback loop, promoting angiogenesis	[165, 166]
Breast Cancer (BC)	Histone H3	H3K18la	KCNK1 → Activates LDHA → Increases glycolysis and induces H3K18la → Upregulates metastasis-related genes (ZWINT, ECT2, etc.) → Reduces cellular rigidity and adhesion.	Promotes tumor cell proliferation, invasion, and metastasis	[167]
Colorectal Cancer (CRC)	Histone H3	H3K18la	lncRNA STEAP3-AS1 → Drives H3K18la → Promotes ERG enrichment at the MMP9 promoter → Degrades the extracellular matrix (ECM).	Promotes tumor invasion and liver metastasis	[168]
Colorectal Cancer (CRC)	VIRMA protein	Protein lactylation	Lactylation enhances VIRMA activity → Increases m6A modification on SP1 mRNA → Elevates SP1 protein levels → SP1 drives TGF-β transcription → Autocrine: enhances EMT; Paracrine: activates ITGA11 + CAFs → CAFs remodel ECM and deliver pro-metastatic circTAX1BP1 via exosomes.	Forms a positive feedback loop driving liver metastasis	[169]

Lactylation as a potential biomarker of tumor endothelial cell functional state

A substantial body of research has elucidated the broad regulatory roles of lactylation in tumor cells. Emerging evidence now indicates that lactylation also functions as a regulator of specific pathological functions in endothelial cells (ECs). It potentially drives aberrant vascular function by modifying key transcription factors, epigenetic regulators, and metabolic enzymes within ECs, demonstrating specificity and functionality that position it as a promising novel biomarker for endothelial cell functional states (Fig. 6).

Fig. 6.

Lactylation drives tumor angiogenesis and metastasis. Lactylation exhibits dual roles in tumor angiogenesis: directly regulating core angiogenic programs, and indirectly shaping the angiogenic niche by promoting invasion and metastasis. This process occurs within the common context of a tumor microenvironment characterized by high lactate levels and hypoxia, which synergistically stabilize HIF-1α and promote two coexisting vascularization modes: endothelial-dependent abnormal vasculature and tumor cell-driven vasculogenic mimicry. The direct pro-angiogenic effects of lactylation are achieved through the following core mechanisms: In hepatocellular carcinoma (HCC), lactylated ABCF1 protein translocates to the nucleus and activates the HIF-1α transcriptional axis, directly upregulating key factors such as VEGF. In glioblastoma (GBM), lactylation directly modifies VE-cadherin, a key structural protein for vasculogenic mimicry, promoting its function and expression. In endometrial cancer (EC), histone H3K18 lactylation activates the classic pro-angiogenic PI3K/AKT/HIF-1α signaling pathway via USP39.The indirect pro-angiogenic effects of lactylation are primarily mediated by driving tumor invasion and metastasis: In breast cancer, lactylation-induced genetic programs reduce cell adhesion and enhance deformability, facilitating tumor cell detachment from the primary site. In colorectal cancer (CRC), lactylation upregulates proteases such as MMP9 to degrade the extracellular matrix, creating physical channels for cell migration. In CRC, lactylation activates cancer-associated fibroblasts via the SP1/TGF-β axis, which remodel the stroma and deliver signaling molecules to establish a pro-metastatic niche. Following successful invasion and colonization of distant organs (e.g., liver metastasis), the subsequent growth of metastatic lesions is inevitably dependent on new blood vessel formation. Therefore, the lactylation-driven invasion-metastasis cascade essentially creates new and persistent “demand scenarios” for angiogenesis

Upon uptake by ECs, tumor-derived lactate can serve as a substrate to drive lactylation of histones in the promoter region of the suppression of tumorigenicity 2 (ST2) gene, significantly upregulating its transcription in tumor-associated endothelial cells. As the membrane receptor for IL-33, ST2 amplifies endothelial sensitivity to IL-33 signaling in the microenvironment [170]. IL-33 is a potent pro-angiogenic molecule, shown in melanoma to upregulate MMP-2/9 via ERK1/2 phosphorylation, promoting vasculogenic mimicry through the IL-33/ST2 axis [171]. Furthermore, the same study suggested that lactylation may inhibit the lymphotoxin-beta receptor (LTβR) signaling pathway, which is crucial for the maturation of high endothelial venules (HEVs), thereby contributing to the structural remodeling of tumor EC plasticity [172]. A similar mechanism has been observed in lung adenocarcinoma, where lactylation at histone H3K14 and H3K18 sites (H3K14la/H3K18la) enriches in the promoter region of the SLC25A29 gene, directly suppressing the transcription of this potentially anti-angiogenic gene and consequently promoting angiogenesis [173]. This metabolite-driven epigenetic silencing leads ECs to exhibit enhanced migration, proliferation, and reduced apoptosis, exemplifying how tumor metabolism reprograms endothelial phenotypes. Lactylation can reinforce communication between tumor cells and ECs. Liang et al., utilizing single-cell spatial transcriptomics, demonstrated that GP73 is highly expressed in both HCC cells and ECs. P300-mediated H3K18 lactylation upregulates GP73 expression. Clusters of HCC cells overexpressing GP73 interact with ECs via multiple angiogenic receptor-ligand pairs (including CXCL5–ACKR1, VEGFA–VEGFR1/R2, TGFB1–(TGFBR1 + TGFBR2), SPP1–(ITGAV + ITGB1), and ANGPTL4–CDH5). Additionally, GP73 elevates the expression of endothelial pro-angiogenic ligands like VEGFA, SPP1, CXCL2, and ANGPTL4, activating the c-MYC/GP73/STAT3 cascade, inducing HUVEC growth and migration in vitro, and promoting HCC angiogenesis in vivo [174]. This reveals that lactylation can enhance intercellular communication and vascular signaling cascades by regulating key intermediary molecules. Correlation with Aggressive Phenotypes and as a Driver: Tumor-specific lactylation modifications correlate directly with a strong pro-angiogenic phenotype, marking a class of ECs highly “educated” and activated by the TME. For instance, in HCC, the level of lactylation at the K312 site of the histone methyltransferase complex component ASH2L (ASH2L-K312la) shows a significant positive correlation with tumor microvessel density. Mechanistically, ASH2L-K312la enhances its chromatin-binding capacity, specifically upregulating VEGF transcription and secretion, driving abnormal proliferation of vascular ECs and mediating hematogenous metastasis [175]. Thus, ASH2L-K312la may serve as a biomarker for driving endothelial proliferative activity. Indicator of Metabolic Dysregulation: Given the strong propensity of tumor ECs for aerobic glycolysis, lactylation is both a product and a consolidator of this metabolic state. In persistently hypoxic environments, the key glycolytic enzyme PKM2 undergoes lactylation in ECs. This modification abnormally stabilizes PKM2 protein by inhibiting its ubiquitin-mediated degradation, exacerbating glycolytic flux and lactate production. This creates a self-amplifying positive feedback loop, leading to severe mitochondrial dysfunction and endothelial metabolic collapse [176]. Consequently, the lactylation level of PKM2 could serve as a crucial indicator of the depth of endothelial metabolic dysregulation. Potential Marker for Thrombosis and Barrier Dysfunction: Interestingly, endothelial lactylation may also be a marker for thrombosis. Specifically, lactate produced via HK2-mediated metabolism in AML cells can be absorbed by ECs, promoting H3K18 lactylation. This enhances PAI-1 expression, ultimately inhibiting clot dissolution and increasing thrombosis risk, revealing a link between endothelial epigenetic modification and coagulation dysfunction [177]. Lastly, endothelial lactylation levels are associated with barrier function. In glioblastoma, temozolomide-resistant tumor cells secrete collagen VI alpha 1 chain (COL6A1), activating the FAK/SRC/Hippo/YAP axis in ECs. This promotes lactylation at the K255 site of the endothelial-specific transcription factor IKZF1 (IKZF1-K255la). Lactylated IKZF1 translocates to the nucleus, driving transcription of genes like UBD, disrupting tight junction integrity in microvascular ECs, reducing blood-brain barrier function, and creating a microenvironment conducive to monocyte-derived macrophage infiltration [178]. As this process depends on paracrine signals from resistant tumor cells, IKZF1-K255la may serve as a specific marker for immunosuppressive and barrier-dysfunctional tumor ECs in resistant GBM. Its detection could potentially be used to assess tumor resistance status and immune microenvironment characteristics.

Lactylation in immune evasion and therapeutic resistance

Within the tumor microenvironment, lactylation not only drives endothelial cell dysfunction to promote abnormal angiogenesis and metastasis as previously described but, more critically, serves as a core immunosuppressive signal that interconnects tumor cells and immune cells [179]. It reinforces the protective “umbrella” effect of tumor vasculature and fosters broad resistance to therapeutic strategies (Table 2).

Table 2.

Mechanisms of lactylation in tumor immune suppression and therapy resistance

Lactylation Site	Cancer Type	Target / Effector	Mechanism	Biomarker Potential	References
Histone Lactylation
H3K18la	NSCLC	POM121 / MYC-PD-L1 axis	H3K18la → POM121↑ → Enhanced MYC binding → PD-L1↑ → Immune evasion	Predictor of immunotherapy response	[180]
H3K18la	HCC	SRSF10 / Glycolysis-M2 Macrophage axis	SRSF10-glycolysis-H3K18la positive feedback loop → Glycolytic enzymes↑ → M2 macrophage polarization	Indicator of immunosuppressive microenvironment	[182]
H3K18la	CRC	Macrophage RARγ / IL-6-STAT3 axis	H3K18la → RARγ↓ → IL-6↑ → STAT3 activation → Pro-tumor M2 phenotype	Marker for tumor-associated macrophage phenotype	[183]
H3K18la	CC	Macrophage GPD2	Lactate from cancer cells → GPD2 (H3K18la)↑ in macrophages → M2 polarization	Marker for malignant progression	[184]
H3K18la	BC	YBX1 / YY1	H3K18la → YBX1/YY1↑ → Enhanced cisplatin resistance	Predictor of chemoresistance	[189]
H3K18la	CRC	RUBCNL / BECN1-Autophagy axis	H3K18la → BECN1 interaction → RUBCNL↑ → Class III PI3K complex recruitment → Hypoxic cell survival↑ → Bevacizumab resistance	Predictor of anti-angiogenic therapy resistance	[19]
H3K18la	HCC	GP73	H3K18la → GP73↑ (Serum levels negatively correlate with lenvatinib response)	Biomarker for monitoring targeted therapy efficacy and prognosis	[174]
H3K9la	HNSCC	IL-11 / JAK2-STAT3 axis	H3K9la → IL-11↑ → JAK2/STAT3 pathway → Exhaustion-related gene activation in CD8⁺ T cells → Immune evasion	Predictor of immunotherapy resistance	[181]
H4K12la	OC	RAD23A	H4K12la → Super-enhancer-mediated → RAD23A↑ → Niraparib resistance	Predictor of PARP inhibitor resistance	[195]
Non-Histone Protein Lactylation	PDAC
ENSA-K63la	BC etc.	STAT3/CCL2 axis	ENSA-K63la → STAT3/CCL2 signaling → TAM recruitment↑ → Immunosuppressive TME	Indicator of immunosuppressive TME extent	[185]
NBS1-K388la	GBM	MRN complex / DNA Repair	NBS1-K388la → MRN complex formation↑ → Homologous recombination repair↑ → Chemoresistance	Predictor of chemosensitivity	[186]
XRCC1-K247la	TNBC	DNA Repair / Nuclear Import	XRCC1-K247la → Nuclear import↑ → DNA repair function↑ → Stemness/radio-chemo resistance	Predictor of radio-/chemo-resistance	[187]
METTL3-K27(de-lactylation)	Glioma	DNA Damage Repair Genes	METTL3 de-lactylation → DNA damage repair genes↑ → Cell survival under cisplatin	Predictor of platinum-based chemoresistance	[189]
PTBP1 (lactylation)	LCSC	PFKFB4 / Glycolysis Cycle	PTBP1 lactylation → Degradation inhibition → RNA-binding capacity↑ → PFKFB4 mRNA stabilization → Glycolytic vicious cycle exacerbation → Stem cell properties maintenance	Marker for cancer stem cell enrichment and recurrence prediction	[190]
ALDOA (lactylation)	GC	DDX17	ALDOA lactylation → DDX17 regulatory function↑ → Stemness maintenance	Marker for stemness	[191]
NSUN2-K508la	HCC	GCLC / Glutathione Synthesis	NSUN2-K508la → Enzyme activity↑ → GCLC mRNA stabilization → GSH↑ → Ferroptosis resistance	Predictor of efficacy for ferroptosis inducers	[192]
PRDX1-K67la	PDAC	NRF2 Antioxidant Pathway	PRDX1-K67la → Nuclear translocation↑ → NRF2 activation → Anti-ferroptotic microenvironment → Regorafenib resistance	Predictor of regorafenib resistance	[196]
TFEB-K91la	Various Cancers	Autophagy/Lysosome Genes	TFEB-K91la → Ubiquitination degradation evasion → Lysosomal activity/autophagy↑ → Cell survival↑	Marker for stress survival capacity	[193]
Vps34 (lactylation)		Beclin1 Complex / Autophagy	Vps34 lactylation → Beclin1 binding↑ → Kinase activity↑ → Autophagy↑ → Therapy resistance	Correlates with autophagy activity and resistance	[194]

Abbreviation: NSCLC (Non-Small Cell Lung Cancer), HCC (Hepatocellular Carcinoma), CRC (Colorectal Cancer), CC (Cervical Cancer), BC (Bladder Cancer), HNSCC (Head and Neck Squamous Cell Carcinoma), OC (Ovarian Cancer), PDAC (Pancreatic Ductal Adenocarcinoma), GBM (Glioblastoma), TNBC (Triple-Negative Breast Cancer), LCSC (Liver Cancer Stem Cells), GC (Gastric Cancer)

Lactylation-mediated immune evasion

Lactylation constructs a multi-layered barrier for immune evasion by targeting immune checkpoints, modulating immune cell function, and shaping an immunosuppressive TME. Upregulation of Immune Checkpoints: In non-small cell lung cancer, histone H3K18 lactylation (H3K18la) activates the transcription of nucleoporin POM121, which in turn enhances the binding of transcription factor MYC to the promoter of the CD274 gene (encoding PD-L1), directly upregulating PD-L1 expression and cloaking tumor cells with an “immunological invisibility cloak” [180]. This epigenetically driven mechanism is not isolated. Research in head and neck squamous cell carcinoma reveals another pathway: specific H3K9 lactylation significantly upregulates interleukin-11 (L-11) expression, which drives the JAK2/STAT3 signaling pathway to activate immune checkpoint genes on CD8⁺ T cells, leading to T cell exhaustion. This process is directly associated with poor patient response to immunotherapy [181]. Reprogramming of Immune Cells: Beyond arming tumor cells themselves, lactylation synergistically suppresses anti-tumor immunity by reprogramming immune cells in the TME. In hepatocellular carcinoma, lactylation and the RNA-binding protein SRSF10 form a vicious cycle: SRSF10 upregulates key glycolytic enzyme expression by stabilizing MYB mRNA, promoting lactate production and H3K18la. H3K18la, in turn, further upregulates SRSF10 and drives macrophage polarization toward an immunosuppressive M2 phenotype [182]. Similarly, in colorectal cancer, H3K18la inhibits the transcription of retinoic acid receptor γ (RARγ) in macrophages, enhancing IL-6 expression and activating the pro-tumor STAT3 pathway, thereby reshaping macrophage function toward a tumor-promoting state [183]. Lactate secreted by cervical cancer cells can also upregulate H3K18la on the GPD2 gene in macrophages, inducing M2 polarization and accelerating malignant progression [184]. Non-histone Modifications: Lactylation of non-histone proteins further reinforces the immunosuppressive network. For example, in pancreatic ductal adenocarcinoma, lactate-induced lactylation of endosulfine alpha (ENSA) at the K63 site triggers the STAT3/CCL2 signaling axis, promoting the recruitment of tumor-associated macrophages to form a robust immunosuppressive niche [185].

Lactylation-driven therapeutic resistance

Concurrently, lactylation fortifies a formidable defense line against various therapies through multiple pathways, including enhancing DNA damage repair, stabilizing pro-survival proteins, maintaining cancer stemness, activating protective autophagy, and altering cell death thresholds. It is a key driver of therapeutic resistance. Enhanced DNA Damage Repair: Lactate promotes lactylation of the DNA repair protein NBS1 at the K388 site, which helps stabilize the MRE11–RAD50–NBS1 (MRN) complex, thereby enhancing homology-directed repair (HDR) of DNA double-strand breaks. This mechanism has been identified in various cancers such as lung, gastric, and breast cancer and is associated with poor prognosis in patients receiving neoadjuvant chemotherapy [186]. Similarly, lactylation of the DNA repair protein XRCC1 at K247 enhances its affinity for nuclear import proteins, promoting its nuclear entry and repair function, which is closely linked to the stemness and radio-/chemoresistance of glioblastoma [187]. Drug Resistance via Transcriptional Regulation: In bladder cancer, upregulation of the transcription factors YBX1 and YY1, driven by H3K18la, is strongly associated with enhanced cisplatin resistance [188]. In triple-negative breast cancer, histone deacetylase 2-mediated delactylation of the methyltransferase METTL3 at K27 helps cells survive cisplatin assault by upregulating DNA damage repair-related genes [189]. Maintenance of Cancer Stemness: Lactylation drives resistance by sustaining the malignant properties of cancer stem cells (CSCs). In glioma stem cells, lactylation of the RNA-binding protein PTBP1 inhibits its TRIM21-mediated degradation, stabilizing PTBP1 and enhancing its binding to PFKFB4 mRNA. This forms a vicious cycle that augments glycolysis and exacerbates tumorigenesis and recurrence [190]. In liver CSCs, lactylated fructose-bisphosphate aldolase A (ALDOA) enhances the stemness-regulating function of DDX17 [191].

Modulation of Cell Death and Autophagy: Lactylation enhances tumor cell survival by regulating cell death and autophagy pathways. In gastric cancer, lactate-mediated lactylation of the methyltransferase NSUN2 at K508 stabilizes the mRNA of the glutamate-cysteine ligase catalytic subunit (GCLC), elevating glutathione levels and thereby increasing cancer cell tolerance to ferroptosis [192]. Lactylation of transcription factor EB (TFEB) at K91 [193] and of the key autophagy kinase Vps34 [194] enhances autophagic activity, helping tumor cells (e.g., in pancreatic ductal adenocarcinoma) cope with therapeutic and microenvironmental stress to promote survival.

Lactylation in resistance to anti-angiogenic and targeted immunotherapies

Crucially, lactylation is directly linked to acquired resistance to various anti-angiogenic or immune-targeted drugs, underscoring its significance as a novel driver of resistance. In ovarian cancer, upregulated H4K12la driven by aberrant glycolysis activates super-enhancer-mediated RAD23A gene expression, promoting resistance to the PARP inhibitor niraparib [195]. In colorectal cancer, tumor tissues from patients resistant to bevacizumab exhibit significantly elevated histone lactylation levels. Mechanistically, H3K18la interacts with the autophagy-related protein BECN1 to promote RUBCNL transcription, facilitating the recruitment of the class III PI3K complex and enhancing the proliferation and survival of hypoxic cancer cells, directly driving bevacizumab resistance [19]. In hepatocellular carcinoma, resistance to regorafenib is closely associated with zinc finger protein 207-driven lactylation of peroxiredoxin 1 (PRDX1) at K67. This modification enhances PRDX1 nuclear translocation and activates the master antioxidant regulator NRF2, creating an anti-ferroptotic, pro-survival microenvironment that allows cancer cells to evade drug-induced death [196]. Similarly, during treatment of HCC with the anti-angiogenic drug lenvatinib, H3K18la was found to be essential for promoting GP73 overexpression. Serum GP73 levels positively correlate with poor patient response to lenvatinib and serve as an independent predictor of treatment efficacy [174]. Collectively, this evidence strongly suggests that targeting lactate production, transport, or specific lactylation modification processes holds great promise as a novel strategy to overcome tumor immune evasion and reverse resistance to anti-angiogenic therapies.

The lactylation regulatory system and its potential as biomarkers

Similar to classical acetylation, protein lactylation is a dynamic and reversible process. This emerging metabolic-epigenetic regulatory mechanism is precisely controlled by three classes of functional molecules: Writers, Erasers, and Readers (Fig. 7). This regulatory axis not only profoundly influences tumor biology but also offers novel perspectives for understanding tumorigenesis, prognosis assessment, and therapeutic monitoring, collectively constituting a “lactyl code” for metabolic signaling in cancer (Table 3).

Fig. 7.

Lactylation modifications drive endothelial cell functional transformation lactylation—a key post-translational modification driven by metabolic rewiring—integratively reprograms endothelial cells by modifying specific histones, transcription factors, and metabolic enzymes. This endows them with multiple pro-tumorigenic phenotypes, including pro-angiogenic, pro-thrombotic, metabolically dysregulated, and barrier-disruptive states, marking a highly activated endothelial cell status “educated” by the tumor microenvironment. Key to symbols: Orange diamonds represent lactate molecules; orange flags represent lactylation modification; ↑ indicates upregulation or enhanced activity; dashed arrows represent indirect effects, migration, or secreted signals; solid arrows represent direct actions or molecular processes

Table 3.

The lactylation code and their biomarker potential

Category	Enzyme/Protein	Cellular localization	Key Substrates/ lactylation sites	Function	Biomarker Potential	Cancer type	References
Writers	p300/CBP	Nucleus	Histone H3K18la	Stimulate TTK and BUB1B transcription to form a glycolysis-H3K18la positive feedback loop, promoting cell cycle progression.	Progression marker, Therapeutic target	PDAC	[197]
	p300/CBP	Nucleus	Histone H3K18la	Upregulate NNMT expression, associated with EGFR-TKI resistance and poor prognosis.	Progression marker, Therapeutic target	NSCLC	[198]
	p300/CBP	Nucleus	Histone H3K18la	Form a positive feedback loop with YBX1 to accelerate tumor progression.	Prognostic/Therapeutic target	HCC	[199]
	p300/CBP	Nucleus	Histone H4K5la	Activate PD-L1 transcription, driving tumor immune evasion.	Immunotherapy target, Response predictor	BLCA	[200]
	p300/CBP	Nucleus	YTHDC1 K82la	Enhance YTHDC1 phase separation ability and stabilizing oncogenic transcripts (BCL2, E2F2) to promote tumor progression.	Prognostic marker	RCC	[201]
	p300/CBP	Nucleus	HNRNPA1 K350la	Promote PKM2 splicing to enhance glycolysis and tumor malignancy.	Prognostic marker	BLCA	[202]
	p300/CBP	Nucleus	Nucleolin K477la	Upregulate MADD via mRNA splicing and activating the ERK pathway to drive tumor growth	Prognostic/Therapeutic target	iCCA	[198]
	AARS1	Cytoplasm	YAP K90 and TEAD1 K108,	Activate the YAP-TEAD complex and promote tumor progression.	Prognostic marker, Therapeutic target	GC	[205, 206]
	AARS2	Cytoplasm	VIRMA K1713la	EnhanceSP1 mRNA stability to promote EMT and liver metastasis.	Metastasis therapeutic target	CRC	[169]
	KAT8 (MOF)	Nucleus	eEF1A2 K408la	Promoting translation elongation and protein synthesis to drive tumor growth.	Prognostic/Therapeutic marker	CRC	[208]
	KAT7 (HBO1)	Nucleus	Histone H3K9la	Promote tumorigenesis	Prognostic marker	CC	[209]
	KAT2A (GCN5)	Nucleus	Histone H3K14/K18la	Promote expression of oncogenic pathways (Wnt/β-catenin) and immune evasion-related genes.	Diagnostic/Prognostic marker	GBM	[210]
	GTPSCS	Nucleus	Histone H3K18la	Promote proliferation and radioresistance	Progression/Therapeutic target	Glioma	[211]
Erasers	HDAC1	Nucleus	Histone H4K12la	Removes acetyl groups from histones	Therapeutic target	CRC	[215]
	HDAC2	Nucleus	Histone H3K18la	Removes acetyl groups from histones	Therapeutic target	PDAC	[197]
	HDAC3	Nucleus	NBS1 K388	reducing DNA repair efficiency and potentially helping to overcome chemotherapy resistance.	Therapeutic target	GC	[217]
	SIRT1	Nucleus and cytoplasm	PKM2 K207la	restore PKM2 enzymatic activity, affecting glycolysis and cell proliferation.	Diagnostic target	HCC	[221]
	SIRT3	Mitochondria	CCNE2 K348la	Exerte a tumor-suppressive effect.	Therapeutic target	HCC	[218]
Readers	TRIM33	Nucleus	Histone H3K18la	Mediate macrophage polarization (M1) to (M2)	Immunoregulatory target	-	[222]
	Brg1 (SMARCA4)	Nucleus	Histone H3K18la	Rgulate transcription and epithelial-mesenchymal transition.	Regulatory target	-	[223]
	DPF2	Nucleus	Histone H3K14la	driving proliferation-related gene transcription, competing with acetylation	Therapeutic target	CC	[224]

Abbreviation: AARS1 Alanyl-tRNA synthetase 1; KAT8 lysine acetyltransferase 8; KAT2A Lysine Acetyltransferase 2 A; LDHA lactate dehydrogenase A; SIRT1 sirtuin 1; CBP CREB-binding protein C; EPB41L4A-AS1 Erythrocyte membrane protein band 41-Like 4 A Antisense RNA 1; GNAT 13 Gcn5-related N-acetyltransferase 13; HDAC1 Histone deacetylase 1; Brg1 Brahma-related gene 1; DPF2 Decreased expression in fat 2

Writers

Lactylation “writers” are enzymes capable of catalyzing the addition of a lactate moiety to specific functional groups (e.g., -OH or -NH2) on target molecules. According to existing research [142], p300/CREB-binding protein (CBP) is considered a key representative lactyltransferase, with p300-catalyzed histone H3 lysine 18 lactylation (H3K18la) being a central event. p300/CBP as Core Writers: In pancreatic ductal adenocarcinoma (PDAC), p300/CBP-catalyzed H3K18la stimulates TTK and BUB1B transcription, forming a glycolysis-H3K18la-TTK/BUB1B-p300 positive feedback loop that promotes cell cycle progression and tumorigenesis, serving as a novel biomarker for PDAC progression [197]. In non-small cell lung cancer (NSCLC), p300/CBP-mediated H3K18la upregulates nicotinamide N-methyltransferase (NNMT) expression, closely associated with EGFR-TKI resistance and poor prognosis in EGFR-mutant NSCLC patients [198]. In hepatocellular carcinoma (HCC), p300/CBP-catalyzed H3K18la forms a YBX1-glycolysis-H3K18la positive feedback loop, accelerating tumor progression [199]. Furthermore, in bladder cancer, p300/CBP-catalyzed H4K5la directly activates the transcription of the immune checkpoint PD-L1, a key driver of immune evasion [200]. Thus, p300/CBP represents a potential therapeutic target, and its detection in tumor tissue may serve as an effective molecular indicator for predicting disease progression, therapy resistance, and immunotherapy response. P300/CBP-mediated lactylation of specific non-histone proteins also holds clear pathological significance. For example, in renal cell carcinoma, p300/CBP catalyzes lactylation of the m6A reader protein YTHDC1 at lysine 82, enhancing its phase separation ability, preventing degradation of oncogenic transcripts BCL2 and E2F2, and promoting cancer cell proliferation, metastasis, and tumor growth [201]. This is correlated with poor prognosis in ccRCC patients. p300/CBP-mediated lactylation of HNRNPA1 at lysine 350 promotes alternative splicing of PKM pre-mRNA towards the PKM2 isoform, enhancing glycolytic flux, proliferation, migration, and invasion in BLCA cells [202]. p300/CBP-catalyzed lactylation of nucleolin at lysine 477 upregulates MAP kinase activating death domain protein (MADD) via precise mRNA splicing and activates the ERK pathway, driving growth of intrahepatic cholangiocarcinoma (iCCA) xenografts and correlating significantly with patient overall survival, suggesting its potential as a prognostic biomarker or therapeutic target in iCCA [198]. Lactyl-CoA serves as the L-lactate donor for lactylation, but its concentration in cancer cells is relatively low (~ 0.011 pmol/10⁶ cells [203] [contrasting with high intracellular lactate levels (10–30 mM [204]). Recent studies indicate that alanyl-tRNA synthetase 1 (AARS1) and AARS2 can directly utilize lactate as a substrate to lactylate various proteins [204]. These enzymes act as lactate sensors and lactyltransferases, binding lactate with micromolar affinity to directly catalyze lysine lactylation [205]. They lactylate multiple proteins (e.g., p53, YAP, cGAS, mitochondrial proteins) [204]. Nuclear-translocated AARS1 lactylates Yes-associated protein (YAP) at K90 and transcriptional enhancer associated domain 1 (TEAD1) at K108, activating the YAP-TEAD complex and forming a positive feedback mechanism with Hippo pathway downstream targets to promote gastric cancer (GC) progression. AARS1 expression is upregulated in GC and correlates with poor patient prognosis [205, 206]. Its mitochondrial homolog, AARS2, promotes lactylation of VIRMA at lysine 1713, enhancing SP1 mRNA stability, mediating TGF-β transcription, and enhancing EMT in CRC cells, presenting a potential therapeutic target for CRC liver metastasis [169].

Other Writer Candidates: KAT family members (e.g., KAT8, KAT2A, KAT7) have been identified as acyltransferases for various histone acylations [207]. High KAT8 expression inversely correlates with overall survival in CRC patients and positively correlates with global Kla levels in CRC tissues. Specifically, KAT8 promotes lactylation of eEF1A2 at K408, enhancing translation elongation and protein synthesis, thereby promoting cancer cell growth and progression; KAT8 knockdown effectively inhibits this process [208]. KAT7 (HBO1) expression increases synchronously with H3K9la in clinical cervical cancer samples, and its inhibition restricts tumor proliferation, migration, invasion, and reduces H3K9la [209]. Recently, KAT2A, by binding phosphorylated ACSS2, has been linked to histone H3K14 and H3K18 lactylation, promoting Wnt/β-catenin, NF-κB, and PD-L1 expression, as well as GBM growth and immune evasion [210]. GTPSCS, a nuclear lactyl-CoA synthetase, is considered important for lactyl-CoA synthesis and histone lactylation. It cooperates with p300 to regulate H3K18la and GDF15 expression, playing a role in glioma proliferation and radioresistance [211].

Erasers

Lactylation “erasers” are enzymes that remove lysine lactylation (Kla) modifications via hydrolysis, restoring the target molecule to its original state. Known erasers include HDACs (histone deacetylases) 1–3 and Sirtuins 1–3, HDACs (1–3): HDACs are the earliest identified delactylases, primarily nuclear-localized and Zn²⁺-dependent. HDAC1-3 are particularly effective in delactylating H4K5 in HeLa cells [212, 213], with HDAC3 exhibiting ~ 1000-fold higher activity than other HDACs in vitro [214]. H4K12la upregulates GCLC expression and inhibits ferroptosis in colorectal cancer stem cells (CCSCs), enhancing chemoresistance; HDAC1 promotes H4 delactylation in CCSCs [215]. Overexpression of HDAC3 reduces NBS1 K388 lactylation levels in gastric cancer under platinum-based neoadjuvant chemotherapy (NAC), decreasing DNA repair efficiency and overcoming chemoresistance [186]. Upregulated HDAC1 and HDAC2 expression correlates with poorer prognosis, and risk models based on them can predict HCC patient outcomes, suggesting their potential as novel HCC biomarkers [216]. Although HDAC2’s role in histone delactylation appears weaker than HDAC1-3, its overexpression can effectively counteract p300-mediated histone lactylation, significantly reducing H3K18la levels in PDAC models [197]. Paradoxically, in hepatic stellate cells (HSCs), pharmacological HDAC inhibition reduces H3K18la levels [217], potentially due to competition between acetylation and lactylation at H3K18. HDAC inhibitors (e.g., Apicidin, MS275) upregulate H3K18ac, leading to downregulation of H3K18la-dependent gene expression and inhibiting HSC activation. These findings suggest that H3K18 serves as a common substrate for acetylation (Kac) and lactylation (Kla), with the final epigenetic effect determined by the dynamic “competition and balance” between multiple modifications. Sirtuins are NAD⁺-dependent deacylases. SIRT3, primarily mitochondrial, mediates delactylation of cyclin E2 (CCNE2) at K348 in HCC; SIRT3 activators exert antitumor effects, marking it as a potential therapeutic target SIRT2 [218], mainly cytoplasmic, efficiently removes lactate groups from synthetic peptides related to PKM2 [219] and exhibits strong delactylase activity at multiple lactylation sites on histones and nucleosomes in vitro [220], suggesting a role in regulating cytoplasmic protein lactylation. SIRT1, primarily nuclear, reduces core histone lactylation upon overexpression in HEK293T cells, while its knockout increases lactylation levels. In HepG2 cells, SIRT1 delactylates PKM2 at K207, restoring its activity and influencing glycolysis and proliferation [221].

Readers

Lactylation “readers” are proteins or domains that recognize and bind the lactate moiety. While definitive lactylation-specific readers are yet to be fully established, several proteins and domains show promise. This bromodomain-containing ubiquitin ligase binds histone PTMs like Kla and Kac with sub-micromolar affinity. It binds histone H3K18la in a sequence-specific manner, mediating the polarization of inflammatory (M1) macrophages towards a reparative (M2) phenotype [222]. The catalytic subunit of the SWI/SNF chromatin remodeling complex can recognize and bind H3K18la, promoting its enrichment at promoters of epithelial function-related genes. This involvement in transcriptional regulation of pluripotency-related genes drives mesenchymal-to-epithelial transition [223]. The double PHD finger 2 protein binds histone H3K14la and co-localizes with it at oncogene promoters in cervical cancer, driving transcription of genes like SEMA5A, ROCK1, and SOAT1 involved in cell proliferation and tumorigenesis [224]. Interestingly, compared to H3K14la, acetylation at the same site marks a significantly different gene set. This suggests that increased H3K14la not only physically reduces H3K14ac but also “hijacks” the regulatory function of the site, switching the downstream gene expression profile from an “acetylation program” to a “lactylation program.” This underscores lactylation as a novel epigenetic signal capable of reprogramming the cellular transcriptome, independent of acetylation.

Tumor lactate/ lactylation-related biomarkers: clinical evidence and translational prospects

Circulating lactate and lactate-related metabolites

Elevated serum lactate levels demonstrate significant differences between patients with various solid tumors (e.g., pancreatic cancer, hepatocellular carcinoma, colorectal cancer) and healthy individuals, indicating its potential as an auxiliary diagnostic biomarker [225]. Furthermore, pre-treatment elevation of serum lactate is significantly correlated with reduced overall survival (OS) and progression-free survival (PFS) in patients with multiple solid tumors, including colorectal and breast cancers, directly reflecting tumor metabolic burden and aggressiveness [226–230]. While a single lactate measurement offers indicative value, metabolite panels or ratios may provide greater specificity. For instance, the lactate-to-pyruvate (L/P) ratio serves as a more sensitive indicator of the intracellular redox state [231]. In pancreatic ductal adenocarcinoma (PDAC), a composite score integrating four metabolites—lactate, pyruvate, citrate, and glucose—demonstrated exceptional discriminatory power (AUC = 0.956) in distinguishing patients from healthy controls, outperforming any single metabolite [225]. This suggests that serum-based metabolomic profiling holds clear translational potential for auxiliary diagnosis and prognostic assessment. Beyond direct metabolite measurement, a multi-gene scoring model has been developed based on the expression of key genes in the lactate metabolism pathway (e.g., a 10-gene signature including LDHA). This score has been validated as a robust independent prognostic indicator in both pan-cancer analyses and specifically in colorectal cancer, and is associated with an immunosuppressive microenvironment and resistance to immunotherapy. Utilizing single-cell RNA sequencing (scRNA-seq) to analyze lactate metabolic activity across different cellular subsets within the tumor microenvironment has revealed that a high lactate metabolism signature is significantly correlated with T cell exhaustion, myeloid-derived immunosuppression, and immunotherapy resistance. These findings provide a molecular basis for stratifying patients likely to benefit from immunotherapy based on metabolic features, while simultaneously pointing toward new strategies for overcoming treatment resistance [232]. Current lactate detection technology is evolving from traditional ex vivo biochemical analysis towards in vivo, non-invasive, and dynamic metabolic imaging. Techniques such as hyperpolarized ¹³C magnetic resonance spectroscopy (HP-¹³C MRS), which enables real-time monitoring of the pyruvate-to-lactate conversion, provide the technical foundation necessary for advancing the clinical translation of lactate detection methodologies [225].

Lactate dehydrogenase (LDH) levels in tumor tissue

Lactate dehydrogenase (LDH), a key enzyme catalyzing the production of lactate—the end-product of glycolysis—serves as a direct indicator of the extent of tumor metabolic reprogramming through its expression and activity. Elevated serum LDH levels are a well-established adverse prognostic factor across various solid tumors and hematologic malignancies (e.g., melanoma, lymphoma, lung cancer) [233] and have been formally incorporated into clinical staging systems for diseases such as advanced melanoma. In melanoma, a meta-analysis confirmed that high serum LDH is significantly associated with shorter overall survival (HR = 1.97, 95% CI 1.62–2.40), with this association remaining robust even in stage IV patients (HR = 1.99, 95% CI 1.59–2.50) [234].

In urologic tumors, preoperative serum LDH levels in bladder cancer patients positively correlate with tumor TNM stage, size, and lymphovascular invasion. Multivariate analysis further identifies LDH ≥ 225 U/L as an independent negative predictor for both overall survival (OS) and progression-free survival (PFS) [235]. Findings in urothelial carcinoma show some variability: a study by Zhang et al. demonstrated that preoperative LDH > 245 U/L is an independent adverse prognostic factor for both OS and disease-free survival (DFS) in multivariate analysis (OS: HR 3.181, 95% CI 1.223–8.276) [236]. In contrast, research by Wei et al. found that preoperative LDH was not an independent prognostic factor for the overall cohort, although within the subgroup with localized disease, elevated LDH (> 220 U/L) was associated with worse OS [237]. This discrepancy may stem from heterogeneity in study populations, sample sizes (e.g., Zhang’s study included only 10 patients with elevated LDH), and the extent of adjustment for confounding variables, underscoring the importance of clinical context when interpreting the prognostic value of LDH.

Beyond static baseline measurements, dynamic changes in LDH levels demonstrate greater prognostic significance in predicting treatment response and survival outcomes. In the context of colorectal cancer liver metastases, research by Zhao et al. showed that patients exhibiting a “Grade 2” increase in serum LDH (where both pre-neoadjuvant chemotherapy and preoperative LDH levels were ≥ the threshold) had poorer pathological responses and a higher incidence of major postoperative complications [238]. This study further identified that preoperative LDH ≥ 231 IU/L was an independent predictor for OS, while pre-neoadjuvant chemotherapy LDH ≥ 145 IU/L was an independent predictor for PFS [239]. Among hepatocellular carcinoma patients treated with atezolizumab plus bevacizumab (ATZ/BEV), those with increasing LDH levels during treatment had significantly shorter PFS compared to those with decreasing levels, and rising LDH was an independent risk factor for poorer PFS in multivariate analysis [240]. Notably, metastatic cancer patients with extremely elevated serum LDH levels (> 1000 IU/L) exhibit a dismal prognosis, with a median OS of only 1.7 months. However, if LDH normalizes within two months post-treatment, their OS can be significantly extended to 22.6 months. This finding provides a critical window for identifying terminal-stage patients and evaluating the efficacy of palliative therapies.

Expression of lactate transporters (MCTs)

Lactate transporters (Monocarboxylate Transporters, MCTs) are key molecules regulating the transmembrane transport of lactate within the tumor metabolic microenvironment. Substantial clinical evidence indicates that the expression levels of MCT1 and MCT4 serve as valuable prognostic biomarkers and hold significant promise as therapeutic targets. Aberrant MCT expression and function have been demonstrated to correlate with poor prognosis in various cancers [241], including melanoma [242] and glioblastoma [243].

Pan-cancer studies have demonstrated that MCT1 and MCT4 are aberrantly overexpressed in various malignancies, and their expression levels are significantly correlated with adverse clinical outcomes in patients. High expression of MCT4 in cancer cells has been widely validated to be associated with shortened overall survival (OS) in multiple solid tumors, including pancreatic [244], breast [245], hepatocellular [246, 247], gastric, and colorectal cancers [248–250]. Similarly, high expression of MCT1 has been established as an independent risk factor for worse progression-free survival (PFS) and OS in head and neck squamous cell carcinoma (HNSCC) patients following chemoradiotherapy, with hazard ratios of 3.1 and 3.8, respectively [251]. In clear cell renal cell carcinoma (ccRCC), high expression of either MCT1 or MCT4 is an independent predictor of shorter PFS, regardless of whether patients receive targeted therapy.

In certain cancers, MCT expression carries unique prognostic significance. In malignant pleural mesothelioma, MCT4 is consistently positive in malignant tissues but absent in normal tissues; its expression is significantly associated with shortened OS, highlighting its potential as both a diagnostic and prognostic marker. In contrast, MCT1 expression lacks clear prognostic value in this context [252, 253]. In T-cell non-Hodgkin lymphoma, overexpression of MCT1 and MCT4 correlates with poor prognosis, and their elevated expression is linked to high-risk clinical features such as advanced disease and high serum LDH levels [254]. In castration-resistant prostate cancer, co-upregulation of MCT4 and GLUT1 is a characteristic feature, serving as a diagnostic tool for identifying aggressive subtypes and a potential therapeutic target [255].

MCT expression is closely intertwined with tumor immune status. Studies indicate that high MCT4 expression negatively correlates with intratumoral CD8⁺ T cell infiltration, suggesting its role in shaping an immunosuppressive microenvironment that compromises anti-tumor immune function [256]. In hepatocellular carcinoma, upregulation of MCT4 expression is associated with an inhibitory immune microenvironment, particularly the differentiation state of Kupffer cells, indirectly contributing to poor prognosis [256]. High MCT4 expression is also closely linked to glycolysis-associated therapy resistance. Selective blockade of MCT4 has been proposed as a promising strategy to overcome such resistance [257].

Small molecule inhibitors targeting MCT1 (e.g., BAY-8002) have shown anti-tumor activity in preclinical studies, particularly in specific cell line models lacking MCT4 expression [258]. However, research has also revealed potential resistance mechanisms. For instance, tumor cells may adapt to MCT1 inhibition by upregulating MCT4 expression or shifting towards oxidative phosphorylation. This suggests that combination strategies, such as co-targeting MCT1 and MCT4 or combining MCT inhibitors with other therapies, may be more effective. Notably, MCT expression is regulated by the chaperone protein CD147, which is a poor prognostic marker in various cancers, including salivary gland carcinoma [259], breast cancer [260], triple-negative breast cancer [261, 262], ovarian epithelial carcinoma [263], and endometrial cancer [264], adding complexity to the targetable pathway.

Detection of lactylation modification

Lactylation, as a direct functional imprint of tumor metabolic reprogramming, demonstrates diagnostic potential supported by pan-cancer analyses. For instance, in gastric cancer, high global lactylation levels independently correlate with larger tumor diameter and vascular invasion [265]. Specific lactylation marks serve as potent, independent prognostic biomarkers, with their expression directly linked to aggressive clinicopathological features and patient survival outcomes.

Among these, H3K18la currently stands out with the most robust clinical correlative evidence. In pancreatic ductal adenocarcinoma (PDAC), elevated H3K18la levels show a significant positive correlation with the severity of perineural invasion (PNI)—an aggressive feature—and are independent predictors of shortened overall survival (OS) and disease-free survival (DFS) [266, 267]. High H3K18la levels are also associated with reduced OS in colorectal cancer patients [19]. Similarly, in breast cancer, high expression of both H4K12la and H4K5la independently correlates with shorter patient OS.

Non-histone lactylation also holds biomarker value. In hepatocellular carcinoma (HCC), lactylation levels at the K430 site of the ABCF1 protein are significantly elevated in tumor tissue, and its high expression independently correlates with poor patient prognosis [157]. In HBV-related HCC, high lactylation levels of AK2 are associated with more aggressive phenotypes, such as tumor thrombus formation, suggesting a poorer prognosis [268].

Beyond direct detection of the modification itself, molecular signatures composed of its upstream regulators or downstream effector genes also possess strong prognostic power. For example, in PDAC, a scoring model based on five lactylation-related genes (including SLC16A1) stratifies patients into subgroups with significant survival differences, and this score is an independent risk factor for OS (HR = 2.297) [269].

The most compelling translational prospect for lactylation lies in its dual potential as a predictive biomarker for treatment response and as a druggable target. For instance, BLM-K24la can predict anthracycline resistance, and intervention strategies based on this finding have entered clinical evaluation (NCT06766266) [266, 267]. In ovarian cancer, LDHA drives both cisplatin resistance and immune escape by promoting both lactate production and histone lactylation (e.g., H3K18la); using the LDHA inhibitor FX11 reduces lactate levels, decreases H3K18la, and restores chemosensitivity [270]. This reveals a universal rationale for targeting lactylation to reverse therapy resistance.

Furthermore, lactylation levels are key regulators of the immune microenvironment. Single-cell evidence indicates that high lactylation or related risk scores associate with an immunosuppressive phenotype, characterized by increased Treg cells, decreased CD8⁺ T cells, and upregulation of checkpoint molecules like PD-L1 [271]. This suggests its potential as a predictive biomarker for immunotherapy benefit. For example, in HCC, patients with low-risk scores show greater sensitivity to immunotherapy [272]. Drug sensitivity analyses further extend its utility for treatment guidance. In colorectal cancer, high-risk patients exhibit greater sensitivity to specific chemotherapeutics (e.g., cisplatin), while low-risk patients may derive more benefit from certain targeted agents [273] (Fig. 8).

Fig. 8.

Clinical biomarkers and translation of tumor lactate/lactylation. Potential clinical biomarkers of tumor lactate metabolism include: (1) circulating lactate and metabolite profiles, which aid in diagnosis and prognosis, with elevated levels correlating with shorter survival; (2) lactate dehydrogenase (LDH), an established prognostic factor whose dynamic changes predict treatment response; (3) monocarboxylate transporters MCT1/MCT4 and their chaperone CD147, where overexpression indicates poor prognosis and immunosuppressive microenvironment; and (4) lactylation modifications (e.g., H3K18la), which serve as independent prognostic biomarkers correlating with tumor aggressiveness and therapy resistance. The clinical translation pathway involves utilizing these biomarkers for diagnosis, prognosis, and treatment monitoring, thereby guiding the development of novel therapies targeting lactate metabolism and lactylation modifications

Druggable targets in anti-cancer therapy: enzymes, transporters, and key sites

As discussed above, lactate and lactylation play crucial roles in tumor angiogenesis and cancer progression. Interventions targeting tumor glycolysis, lactate metabolism, and lactylation modifications have demonstrated significant therapeutic potential across multiple cancer types (Table 4), positioning this field as a prominent focus in cancer research. Importantly, targeting the vulnerabilities in lactate metabolism not only undermines the metabolic advantages of tumor cells but also holds promise for indirectly or directly suppressing tumor angiogenesis. This approach synergizes with immunotherapy and anti-angiogenic therapy, offering a strategic opportunity for a combined “metabolic-immuno-vascular” intervention framework (Fig. 9).

Table 4.

Onco-therapeutic drugs targeting lactate metabolism/lactylation

Name	Structural Formula	Target / Key Lactic-Modified Site	Mechanism of Action	Cancer Type	Clinical Trial Stage	Major Toxicity / Concerns	References
Plumbagin		GLUT-1	Inhibits GLUT-1 and VEGF-A expression, blocking angiogenesis.	Ovarian Cancer	Preclinical	Dose-dependent hepatotoxicity and reproductive toxicity in animal studies	[275]
Apigenin		GLUT-1	Antagonizes HIF-1α-mediated upregulation of GLUT-1 and VEGF.	Pancreatic Cancer	Preclinical	Generally recognized as safe (GRAS)	[276]
AZD6244 (Selumetinib)		GLUT-1	MEK1/2 inhibitor that downregulates GLUT-1 and VEGF via ERK pathway inhibition.	General solid tumors	Preclinical	Rash, diarrhea, nausea, vomiting, fatigue, peripheral edema, Ocular toxicity	[277]
3PO		PFKFB3	Inhibits glycolysis in tumor endothelial cells, promotes vascular normalization (stabilizes VE-cadherin, increases pericyte coverage).	Glioblastoma	Preclinical	Not fully characterized	[315]
Shikonin		PKM2	Inhibits PKM2 SUMOylation/dimerization, downregulates β-catenin/VM-related genes, reverses CSC stemness.	Ovarian Cancer	Preclinical	Not fully characterized	[280]
AT-101 (Gossypol aceticate)		LDHA	Competes with NADH at the binding site of LDHA.	NSCLC, Head and Neck Cancer, Prostate Cancer, Gastroesophageal Cancer	Phase I/II	Dose-limiting gastrointestinal toxicity (nausea, diarrhea), fatigue, hypokalemia. Male reproductive toxicity	[284]
FX-11, Galloflavin, N-hydroxyindoles		LDHA	Competitively inhibit LDHA.	Colorectal Cancer, Melanoma, Breast Cancer	Preclinical	Not fully characterized	[284]
Oxamate		LDHA	Pyruvate analog, competitively inhibits LDHA.	Melanoma, NSCLC	Preclinical	Not fully characterized	[286, 287]
MS6105 (NCI006)		LDHA/B	Recruits VHL E3 ligase to degrade LDHA/B.	Pancreatic Cancer	Preclinical	Not fully characterized	[288]
AZD3965		MCT1	Selective oral inhibitor of MCT1, blocks lactate export.	Breast Cancer Advanced Solid Tumors & Lymphoma	Phase I (NCT01791595)	Dose-limiting ocular toxicity (asymptomatic, reversible)	[292]
AZ93	-	MCT4	Selective inhibitor of MCT4.	Colorectal Cancer	Preclinical	Not fully characterized	[293]
7-Aminocarboxycoumarin (7ACC)		MCT4	Inhibits mitochondrial pyruvate carrier (MPC), reduces lactate production, and attenuates MCT4 compensation.	Cancer metabolism	Preclinical	Not fully characterized	[294, 295]
AC-73, Metuzumab		CD147 (MCT1/4 chaperone)	Inhibit CD147 function, impairing membrane localization of MCT1/4.	Prostate Cancer	Preclinical	Not fully characterized	[296]
DML	-	Histone Lactylation (H3K9la, H3K56la)	Inhibits specific histone lactylation sites.	Hepatocellular Carcinoma (HCC)	Preclinical	Not fully characterized	[298]
Royal Jelly Acid (RJA)		Histone Lactylation (H3K9la, H3K14la)	Inhibits specific histone lactylation sites.	Hepatocellular Carcinoma (HCC)	Preclinical	Not fully characterized	[299]
C646		Histone Lactylation (broad, via p300 inhibition)	Broad-spectrum histone acetyltransferase (p300) inhibitor, also suppresses histone lactylation.	HCC and other cancers	Preclinical	Not fully characterized	[164]
K673-peptide (K673-pe)	-	MRE11 Lactylation (K673)	Cell-penetrating peptide that specifically blocks MRE11 lactylation.	Colon Cancer	Preclinical	Not fully characterized	[300]
D34-919	-	PKM2 (tetramerization)	Inhibits PKM2 tetramerization and activity, reducing lactate production and downstream effects (e.g., XRCC1 lactylation).	Glioblastoma	Preclinical	Not fully characterized	[301]
β-Alanine		p53 Lactylation (K120/K139)	Reduces lactylation of p53, restoring its transcriptional activity.	General tumor models	Preclinical	Generally safe as a dietary supplement. High doses may cause paresthesia (tingling sensation).	[206]
MG149		KAT8 (lactylation writer)	HDAC inhibitor that downregulates KAT8, blocking the KAT8-eEF1A2-Kla axis.	Colorectal Cancer	Preclinical	Not fully characterized	[208]
Honokiol		CCNE2 Lactylation (via SIRT3 activation)	Natural product that activates the de-lactylase SIRT3.	Hepatocellular Carcinoma (HCC)	Preclinical	Generally well-tolerated in preclinical and limited human studies. Potential for bleeding risk	[218]
SAHA (Vorinostat), TSA (Trichostatin A)		HDAC1 Lactylation (K412la) / FTO/ALKBH5	HDAC inhibitors that reduce HDAC1 lactylation, activating FTO/ALKBH5 to degrade FSP1 mRNA and promote ferroptosis.	Colorectal Cancer	Preclinical	Fatigue, diarrhea, nausea, thrombocytopenia, anemia, QTc prolongation.	[302]
Tubuloside A		ABCF1 Lactylation (K430la)	Specifically inhibits lactylation at K430 of ABCF1.	Hepatocellular Carcinoma (HCC)	Preclinical	Not fully characterized	[153]
Irinotecan		BLM Lactylation	May inhibit lactylation of BLM helicase, restoring sensitivity to anthracyclines.	Resistant Bladder Cancer	Phase I (NCT06766266)	Severe diarrhea (early & delayed), myelosuppression (neutropenia), nausea/vomiting, fatigue.	[303]

Abbreviations: 2-DG, 2-Deoxy-D-glucose; ABCF1, ATP-binding cassette sub-family F member 1; AE, Adverse Event; BLM, Bloom syndrome protein; CBP, CREB-binding protein; CCNE2, Cyclin E2; CD147, Cluster of Differentiation 147; CPK, Creatine Phosphokinase; CSC, Cancer Stem Cell; CTCL, Cutaneous T-cell Lymphoma; eEF1A2, Eukaryotic translation elongation factor 1 alpha 2; GLUT, Glucose Transporter; GRAS, Generally Recognized As Safe; HDAC, Histone Deacetylase; HDACi, HDAC Inhibitor; HIF-1α, Hypoxia-Inducible Factor 1-alpha; HR, Homologous Recombination; ICB, Immune Checkpoint Blockade; Kla, Lysine Lactylation; LDHA, Lactate Dehydrogenase A; MCT, Monocarboxylate Transporter; MPC, Mitochondrial Pyruvate Carrier; mRNA, Messenger RNA; NF1, Neurofibromatosis Type 1; NRF2, Nuclear factor erythroid 2–related factor 2; NSCLC, Non-Small Cell Lung Cancer; PARPi, Poly (ADP-ribose) Polymerase Inhibitor; PFKFB3, 6-Phosphofructo-2-kinase/fructose-2,6-bisphosphatase 3; PKM2, Pyruvate Kinase M2; PROTAC, Proteolysis-Targeting Chimera; PRDX1, Peroxiredoxin 1; ROS, Reactive Oxygen Species; RPED, Retinal Pigment Epithelial Detachment; SIRPα, Signal-regulatory protein alpha; SIRT3, Sirtuin 3; siRNA, Small interfering RNA; SUMO, Small Ubiquitin-like Modifier; TKI, Tyrosine Kinase Inhibitor; TME, Tumor Microenvironment; TVN, Tumor Vascular Normalization; VEGF, Vascular Endothelial Growth Factor; VEGFR2, Vascular Endothelial Growth Factor Receptor 2; VHL, Von Hippel-Lindau; VM, Vasculogenic Mimicry; XRCC1, X-ray Repair Cross Complementing 1; ZNF207, Zinc Finger Protein 207

Fig. 9.

Multidimensional therapeutic strategies targeting tumor lactate metabolism and lactylation modifications. This schematic illustrates three synergistic therapeutic strategies. First, targeting metabolic vulnerability (blue pathway) involves inhibiting lactate production and efflux using small-molecule inhibitors against key enzymes and transporters such as GLUT1, PFKFB3, PKM2, LDHA, and MCT1/4 (e.g., AZD3965/AZ93). Second, reversing immunosuppression (red pathway) aims to reduce lactate levels in the tumor microenvironment, which restores T and NK cell function and decreases Treg and M2-like TAM recruitment, thereby enhancing the efficacy of immunotherapies. Third, promoting vascular normalization (purple pathway) combines glycolysis inhibitors with anti-angiogenic agents (e.g., bevacizumab) to lower lactate, counteract acidosis, and inhibit lactylation-driven pro-angiogenic signals, ultimately improving oxygenation and drug delivery. The triangular model at the base highlights that simultaneously targeting these interconnected metabolic, immune, and vascular pathways can produce synergistic effects, overcome single-agent resistance, and provide a framework for effective combination therapies

Targeting lactate metabolic enzymes

Glucose transporters (GLUTs) mediate the influx of glucose into cells. Subsequent catalysis by key glycolytic enzymes such as PFKFB3 and PKM2 generates pyruvate, the direct precursor to lactate [274], positioning GLUTs as upstream regulators of lactate production. In ovarian cancer, the natural compound plumbagin inhibits angiogenesis by suppressing VEGF-A and GLUT-1 expression, thereby blocking pro-angiogenic signaling in human umbilical vein endothelial cells (HUVECs) and inhibiting endothelial proliferation and tumor vascularization [275]. Similarly, the flavonoid apigenin exerts anti-proliferative and anti-angiogenic effects in pancreatic cancer by antagonizing the HIF-1α-mediated upregulation of GLUT-1 and VEGF [276]. Furthermore, the MEK1/2 inhibitor AZD6244, by inhibiting the ERK signaling pathway, downregulates GLUT-1 and VEGF expression, significantly reducing tumor vascular perfusion. This suggests that targeting GLUTs can synergistically enhance anti-angiogenic therapy [277].

PFKFB3 is a key allosteric activator of glycolysis and plays a central role in the abnormal vasculature of tumors via its activity in tumor endothelial cells. Inhibiting PFKFB3 does not significantly affect tumor cell growth per se but exerts therapeutic effects by promoting vascular normalization. The underlying mechanisms include stabilizing VE-cadherin to enhance the vascular barrier, upregulating N-cadherin to promote pericyte coverage, and inhibiting tumor cell-endothelial adhesion via the NF-κB pathway, thereby effectively reducing tumor invasion and metastasis [278]. This provides a novel therapeutic rationale for targeting the tumor metabolic microenvironment rather than the tumor cells themselves.

PKM2 is a rate-limiting enzyme in the final step of glycolysis. Our previous research found that shikonin (SHK) inhibits SUMOylation and dimer formation of PKM2 [279], downregulates the β-catenin signaling pathway and vasculogenic mimicry (VM)-related genes, thereby reversing cancer stem cell (CSC) stemness, inhibiting VM formation, and blocking alternative nutrient supply to tumors [280].

LDH catalyzes the reversible conversion between pyruvate and lactate [281]. High LDHA expression is closely associated with poor patient prognosis, making it a significant therapeutic target [282]. Various LDHA inhibitors have shown potential in preclinical and clinical studies [283]. Gossypol acetic acid (R-(-)-AT-101) demonstrated efficacy in phase I/II trials for advanced NSCLC, head and neck cancer, and prostate cancer. Low-dose AT-101 (≤ 30 mg/day) combined with chemoradiotherapy achieved high complete response rates and long-term survival benefits in patients with gastroesophageal cancer [284]. Its derivatives, such as FX-11, galloflavin, and N-hydroxyindole compounds, inhibit tumor growth in colorectal cancer, melanoma, and breast cancer models by competitively inhibiting the NADH binding site [285]. Oxamate, a pyruvate analog, competitively inhibits LDHA, leading to pyruvate accumulation and reduced lactate production. In melanoma models, oxamate combined with metformin synergistically inhibits tumor growth [286]. In NSCLC, LDHA inhibition by oxamate causes reactive oxygen species (ROS) accumulation and defective DNA damage repair, thereby enhancing radiosensitivity [287]. Furthermore, novel degraders such as MS6105 (NCI006), which recruit the VHL ubiquitination system to specifically degrade LDHA/B, effectively inhibit lactate production and tumor proliferation in pancreatic cancer models [288].

Targeting lactate transporters (MCTs)

MCTs represent the most extensively studied lactate transporters in current cancer research. Both MCT1 and MCT4 require association with the glycoprotein chaperone CD147 (basigin) for proper membrane localization and functional expression [289]. Studies indicate that when MCT1 function is inhibited, tumor cells can functionally compensate by upregulating MCT4 expression, thereby maintaining lactate efflux and leading to drug resistance [290]. Consequently, strategies to simultaneously target both MCT1 and MCT4, or to block their common chaperone CD147, have become crucial for overcoming compensatory resistance. AZD3965, an orally available selective MCT1 inhibitor, demonstrated promising anti-tumor activity and an acceptable safety profile in preclinical breast cancer models [291]. Its Phase I clinical trial (NCT01791595) in patients with advanced solid tumors (including NSCLC, gastric, and breast cancers) and B-cell lymphoma has been completed. The results indicate that AZD3965 was well-tolerated within the target dose range, with dose-limiting toxicities primarily consisting of asymptomatic and reversible ocular changes [292]. This study provides preliminary safety evidence supporting the clinical translation of MCT1 inhibitors. Considering compensatory upregulation of MCT4, selective MCT4 inhibitors such as AZ93 have been developed. In colorectal cancer models, the combination of AZ93 with AZD3965 synergistically inhibited tumor cell proliferation, outperforming either agent alone [293]. Furthermore, compounds like 7-aminocarboxycoumarin (7ACC), which inhibit the mitochondrial pyruvate carrier (MPC), can block mitochondrial pyruvate metabolism, thereby suppressing lactate production and weakening MCT4’s compensatory function. This offers a novel approach for combination metabolic therapy [294, 295]. CD147, as an essential chaperone for MCT1/4 membrane expression, is another viable target. Inhibitors such as AC-73 and the monoclonal antibody metuzumab can reduce lactate efflux and inhibit prostate cancer progression by interfering with CD147 function [296]. However, because CD147 is involved in diverse cellular processes (e.g., cell adhesion, protease activation), targeting this protein requires precise therapeutic window design to minimize off-target toxicity. Recent research demonstrates that inhibiting lactate transport can effectively improve immune cell function and enhance the efficacy of immune checkpoint blockade (ICB). In a preclinical colorectal cancer model, the selective MCT4 inhibitor MSC-438,138 reduced tumor lactate secretion and improved T cell function. When combined with an anti-PD-L1 antibody, it significantly promoted tumor cell lysis, increased T cell infiltration and functionality in vivo, and decreased the frequency of inhibitory myeloid cells [256]. Further studies confirmed that inhibiting MCT4 alone could raise intratumoral pH, enhance leukocyte infiltration and T cell activation, and synergize with anti-PD-1 therapy to delay tumor growth and prolong survival. Interestingly, combined inhibition of MCT1 did not provide additional benefit [297]. This suggests that MCT4 may play a dominant role in mediating lactate-driven immunosuppression, and its selective inhibition represents a promising immunometabolic modulation strategy.

Targeting lactylation

Current research on drugs targeting histone lactylation has shown initial promise. For example, demethylzeravone (DML) inhibits lactylation at H3K9 and H3K56, thereby suppressing tumor progression in hepatocellular carcinoma (HCC) models [298]. Xu et al. discovered that royal jelly acid (RJA) inhibits HCC development by suppressing histone lactylation at the H3K9 and H3K14 sites [299]. The broad-spectrum histone acetyltransferase inhibitor C646 has also been shown to significantly inhibit histone lactylation and effectively suppress tumor growth in various preclinical models, including HCC [164].

Lactylation of non-histone proteins, which is extensively involved in DNA damage repair, transcriptional regulation, and metabolic pathways, is also emerging as a novel therapeutic target. For instance, the homologous recombination (HR) repair protein MRE11 undergoes lactylation at K673, mediated by the acetyltransferase CBP. This modification promotes DNA end resection and HR repair, driving tumor progression. The cell-penetrating peptide K673-pe, which specifically targets this site, blocks MRE11 lactylation, inhibits HR activity, and enhances the sensitivity of colon cancer cells to cisplatin and PARP inhibitors [300]. Furthermore, tetramerization of pyruvate kinase M2 (PKM2) promotes lactate production, which in turn mediates the lactylation and subsequent nuclear translocation of the DNA repair protein XRCC1, leading to chemotherapy resistance [301]. The small molecule inhibitor D34-919, which inhibits PKM2 tetramerization, can sensitize glioblastoma cells to radiotherapy [187]. Lactylation of the tumor suppressor p53 at its DNA-binding domain (K120/K139) inhibits its transcriptional activity and promotes tumorigenesis. Beta-alanine can reduce p53 lactylation, thereby attenuating tumor growth in animal models [206].

The lactylation enzyme system itself also presents therapeutic opportunities. KAT8-mediated lactylation of eEF1A2 supports the high translational demands of tumors. The HDAC inhibitor MG149 inhibits colorectal cancer progression by downregulating KAT8 and blocking this signaling axis [208]. Additionally, the natural compound honokiol activates SIRT3, inducing the delactylation of cyclin CCNE2 and inhibiting HCC growth [218], suggesting the therapeutic potential of SIRT3 agonists. Research has found that lactylation at lysine 412 (K412la) of HDAC1 is crucial for ferroptosis resistance in colorectal cancer cells. HDAC inhibitors such as vorinostat (SAHA) and trichostatin A (TSA) can activate the demethylases FTO/ALKBH5, reducing m⁶A modification on the mRNA of the ferroptosis suppressor protein FSP1 and promoting its degradation, thereby enhancing ferroptosis sensitivity [302]. The development of small molecule inhibitors targeting specific lactylation modifications has progressed. For example, Tubastatin A specifically inhibits lactylation at lysine 430 (K430la) of the transporter ABCF1, showing anti-tumor activity in preclinical HCC models [157]. One of the most translationally promising studies focuses on the lactylation of the DNA helicase BLM, which is associated with tumor resistance to anthracyclines. An ongoing Phase I clinical trial (NCT06766266) is evaluating the safety and feasibility of combining liposomal irinotecan (which may block lactylation) with epirubicin for the treatment of drug-resistant bladder cancer [303]. The development of these drugs signifies that therapeutic strategies targeting lactylation are gradually transitioning from basic research to clinical validation.

Adjuvant immunotherapy

Lactate is a key tumor immunosuppressive factor, and targeting lactate metabolism may resensitize immunologically evasive tumors. Li et al. found that the MCT-4 inhibitor Syrosingopine enhances immunotherapy by reducing lactate efflux, increasing the polarization of M1-like tumor-associated macrophages (TAMs), elevating the proportion of natural killer (NK) cells, and reducing the number of regulatory T cells (Tregs). This validates a novel strategy of modulating lactate metabolism to synergize with tumor immunotherapy in breast cancer [304]. Furthermore, combining lactate dehydrogenase A (LDHA) inhibition with CAR-T cell therapy can also enhance immunotherapeutic outcomes. H3K18la upregulates the activity of promoters for genes encoding the ectonucleotidases CD39, CD73, and the chemokine receptor CCR8 in CD4 + T cells and macrophages within lactate-rich glioma stem cell niches. Oxamate, as an LDH inhibitor, reduces lactate production and lowers H3K18la, preventing CAR-T cell exhaustion. Concurrently, it decreases the expression of PD-1 and Tim-3 on CAR-T cells, thereby improving their efficacy against glioblastoma [305]. Recent research has identified lactate oxidase (LOx)-based nanocatalytic therapy as another breakthrough for enhancing immunotherapy. This approach reduces lactate concentration and releases immunostimulatory hydrogen peroxide, boosting the efficacy of immune checkpoint inhibitors while mitigating tumor immunosuppression [306]. In a breast cancer model, encapsulating LOx and a gene-editing plasmid for signal regulatory protein alpha (SIRPα) within metal-organic frameworks (MOFs) induced CD47-SIRPα blockade through lactate depletion, effectively enhancing the anti-tumor activity of macrophages [307]. Additionally, newly reported targeting strategies aim to block GLUT1, whose expression can inhibit tumor lactate efflux and significantly alleviate lactate-driven immunosuppression. In glioblastoma (GBM), intratumoral injection of BAY-876, a GLUT1 inhibitor, delayed tumor progression and extended survival [308]. In melanoma cells, combining the lactate dehydrogenase inhibitor GSK2837808A with adoptive T cell therapy (ACT) may enhance therapeutic effects and reduce tumor drug resistance [309]. In hepatocellular carcinoma (HCC), the SRSF10/glycolysis/H3K18la positive feedback loop is associated with immunotherapy resistance. Using the selective inhibitor 1C8 to target SRSF10 may inhibit M2 macrophage polarization and potentiate the therapeutic effect of PD-1 monoclonal antibodies in tumors [182]. Therefore, research aimed at enhancing immunotherapy efficacy by modulating lactate metabolism in the TME holds substantial promise.

Adjuvant anti-angiogenic therapy

Anti-angiogenic therapy, which inhibits tumor progression by blocking blood supply, has become a cornerstone of cancer treatment [310]. However, prolonged use can induce adaptive changes in tumors, such as metabolic reprogramming and compensatory signaling activation, ultimately leading to abnormal vascular regeneration, increased metastatic risk, and treatment resistance [4]. Tumor vascular normalization (TVN) aims to remodel tumor vascular function, alleviate hypoxia, enhance immune infiltration, and improve drug delivery efficiency, offering a novel approach for combination therapyn [311–313]. Yet, the therapeutic window for TVN is transient, and single-agent effects are limited. Therefore, combining anti-angiogenic agents with metabolic modulators to achieve synergistic effects has emerged as a new strategy to enhance clinical benefit [314].

As emphasized earlier, lactate plays a pivotal role in angiogenesis. Targeting lactate metabolism holds promise for multi-dimensionally inhibiting pro-angiogenic stimuli, thereby extending and stabilizing the TVN window and optimizing anti-angiogenic therapeutic outcomes.

Inhibiting lactate production to synergistically induce vascular normalization

Studies show that combining anti-VEGF agents with glycolysis inhibitors enhances TVN effects. For example, in orthotopic glioblastoma models, the combination of bevacizumab and the PFKFB3 inhibitor 3PO downregulated the key TVN regulator Tie1, alleviated hypoxia, reduced lactate production, and significantly improved the delivery and efficacy of chemotherapeutic drugs. Imaging confirmed that this combination promoted global improvement in tumor vascular function [315]. Similarly, co-delivering bevacizumab and the glycolysis inhibitor dichloroacetate within nanoparticles simultaneously inhibited angiogenesis and counteracted the lactate elevation induced by monotherapy, demonstrating synergistic therapeutic potential [316]. Furthermore, in renal cell carcinoma, the natural compound zingerone A inhibited LDHA, reduced lactate production, and disrupted the HIF-1α/VEGF/VEGFR2 signaling axis. This significantly enhanced the efficacy of sunitinib in both sensitive and resistant models, providing a novel metabolic intervention pathway to overcome TKI resistance [317].

Reversing the acidic microenvironment to restore drug sensitivity

Acidification of the tumor microenvironment can impair the efficacy of tyrosine kinase inhibitors (e.g., sunitinib). Alkalinizing therapy (e.g., co-administration of sodium bicarbonate) can restore tumor vascular responsiveness to sunitinib, improving its anti-angiogenic effects [153].

Intervening in lactylation modifications to block pro-angiogenic signaling

In colorectal cancer, using glycolysis inhibitors (2-deoxy-D-glucose [2-DG] and oxamate) or targeting LAMC2 with siRNA to reduce histone lactylation levels (e.g., H3K18la) enhanced the efficacy of bevacizumab [229]. Additionally, siRNA targeting SEMA6A disrupted the upstream lactylation-related SEMA6A-Rho axis, eliminating its pro-angiogenic effect [318]. Regarding resistance, ZNF207-mediated lactylation of peroxiredoxin PRDX1 at K67 is a key mechanism of regorafenib resistance in hepatocellular carcinoma; interfering with PRDX1 lactylation or its downstream NRF2 activity can reverse this resistance [196].

Combining with immunotherapy to construct a synergistic therapeutic network

Anti-angiogenic therapy can improve the immune microenvironment and holds synergistic potential with immune checkpoint inhibitors. Research shows that anti-VEGF combined with anti-PD-L1 therapy extends the TVN window and enhances CD8⁺ T cell function [319]. Given that lactate drives both immunosuppression and angiogenesis, targeting lactate metabolism has the potential to form a three-dimensional synergy with “anti-angiogenic + immunotherapy.” This approach could simultaneously alleviate immunosuppression and vascular abnormalities, thereby maximizing efficacy and delaying resistance. In rectal cancer, a triple therapy combining a lactylation inhibitor with bevacizumab and immunotherapy showed significant activity in resistant models [319]. The natural anti-cancer compound evodiamine inhibits HIF-1α expression, thereby attenuating the suppression of Sema3A expression, limiting H3K18la histone lactylation and PD-L1 expression in prostate cancer cells, and multi-effectively blocking lactate-induced angiogenesis [320].

Current challenges in lactylation research

Despite significant advancements in lactylation (Kla) research, it is essential to recognize that fundamental challenges persist, impeding the field’s transition from descriptive observations to mechanistic understanding and from correlation to causation.

Ambiguity of molecular mechanisms and tissue specificity

The core challenge in Kla research lies in the incomplete elucidation of its molecular basis. Firstly, the causal relationship of Kla occurrence remains unclear: Is it an active, programmatically regulated signaling event, or a passive metabolic byproduct of elevated lactate? Secondly, while the “writer” and “eraser” enzyme systems responsible for dynamic Kla regulation have been partially identified, their substrate specificity, catalytic kinetics, and regulatory principles across different cellular contexts are poorly understood. Notably, these regulatory enzymes exhibit significant expression and functional heterogeneity across cancer types. For example, p300/CBP functions as a “writer” of histone lactylation in multiple cancers [197, 321], whereas TIP60 specifically catalyzes lactylation of the NBS1 protein to promote DNA repair in prostate cancer [186]. Similarly, “erasers” like HDAC3 and HDAC2 demonstrate cancer type-specific substrate preferences [189]. This tissue-specificity indicates that the functions of lactylation are highly dependent on tumor context, posing a significant challenge to the generalizability and reproducibility of research findings.

Complexity of the tumor microenvironment and temporal dynamics

The heterogeneity of the tumor microenvironment (TME) dictates that different cell types (cancer cells, immune cells, stromal cells) exhibit distinct sensitivities and response mechanisms to lactate. A key unresolved question is whether lactylation events follow a specific temporal sequence within the TME. Do cancer cells, as primary lactate producers, undergo lactylation first, subsequently influencing other cells by altering the microenvironment? Alternatively, do tumor stromal cells experience lactylation at an early stage, thereby impacting cancer cells? Dynamic tracking studies are required to validate these possibilities.

Crosstalk with other post-translational modifications

Lactylation is not an isolated event; it engages in complex crosstalk with other crucial post-translational modifications (PTMs), such as acetylation and ubiquitination [322]. This interaction manifests primarily in two ways: Site Competition: On identical lysine residues of proteins like p53, histone H3, and PKM2, lactylation, acetylation, and ubiquitination may be mutually exclusive. The occurrence of lactylation might directly preclude other functionally critical modifications [206]. Shared and Competing Enzyme Activities: Certain deacetylases (e.g., HDAC3, SIRT2) also possess delactylase activity, and “writer” enzymes like AARS1 and p300 may compete for the lactyl-CoA substrate pool [213]. This functional overlap among enzymes implies that fluctuations in lactylation levels can directly impact the homeostasis of traditional modifications like acetylation.

Intrinsic limitations of research methodologies

Current experimental approaches for studying Kla have inherent limitations that may compromise the physiological relevance of conclusions. In vitro studies often employ supraphysiological concentrations of lactate or lactyl-CoA to induce observable phenotypes, potentially amplifying the biological role of Kla beyond its relevance in actual pathological states. Common interventions, such as lactate supplementation or LDHA inhibition, not only alter Kla levels but also trigger widespread changes in redox status and metabolic networks. Consequently, distinguishing the specific effects of Kla from its indirect consequences in vivo remains challenging.

Bottlenecks in pathophysiological function studies

Functionally, although Kla is frequently associated with processes like tumorigenesis and immune evasion, its precise causal mechanisms and downstream signaling networks remain obscure. For instance, the specific role of Kla in epigenetic regulation—including how it is selectively recruited to specific genomic loci and how it influences target gene transcription—awaits comprehensive characterization. Technologically, we lack tools for the precise, dynamic modulation of intracellular Kla levels within a physiological range. Existing genetic and pharmacological interventions are often accompanied by significant off-target effects, hindering the establishment of direct causal relationships between Kla and phenotypic outcomes.

Conclusion and future perspectives

In summary, this review systematically delineates the central role of lactate metabolism and its mediated lactylation modification in regulating tumor angiogenesis. Key molecules, represented by lactate metabolic enzymes, transporters, and specific lactylation sites, collectively constitute a dynamic and targetable “metabolic-epigenetic” regulatory axis. Targeting this axis not only directly disrupts the tumor’s “Warburg effect,” cutting off its energy and biosynthetic precursor supply, but also, by modifying histone and non-histone functions, reshapes the immunosuppressive tumor microenvironment and promotes the normalization of the structurally and functionally aberrant tumor vasculature. This multi-faceted approach holds significant therapeutic promise.

However, the field faces several major bottlenecks, including the limitations of lactylation detection technologies, questionable physiological relevance of in vitro high-lactate conditions, challenges in reproducibility across different tumor types, and the impact of cellular heterogeneity. To advance the field, future efforts should focus on the following key directions: (1) Methodological Innovation: Develop site-specific lactylation detection tools and probes capable of real-time tracking of lactylation dynamics to overcome current methodological limitations. (2) Functional Screening: Utilize CRISPR-based high-throughput screening platforms to systematically identify key “writers,” “readers,” and “erasers” that regulate lactylation in diverse cancer contexts. (3) Pharmacological Intervention: Develop and employ specific small-molecule inhibitors or activators to modulate lactylation flux in preclinical models, while ensuring the physiological relevance of these interventions. Elucidating the precise molecular architecture and functional networks of the lactylation pathway will ultimately map the complete regulatory landscape of “lactate metabolism — lactylation modification — vascular phenotype.” Based on this map, future therapeutic strategies will evolve towards greater sophistication and systematicity. By designing well-orchestrated, sequential, and combinatorial treatment regimens integrating “metabolic intervention — immunomodulation — vascular normalization,” and by using liquid biopsy- or radiomics-based biomarkers for precise patient stratification, we may fundamentally overcome tumor heterogeneity and adaptive resistance. This approach offers a promising innovative pathway towards achieving long-term functional cures for cancer.

Author contributions

PT Wu, J Zhang wrote the paper, ZW Jiang, XR Liu. J Zhou and N Jiang draw pictures, X Chen, DC Wu, YK Li designed the project and revised the paper. All authors contributed to the article and approved the submitted version.

Funding

The present study was supported by the Natural Science Foundation of China (82303246), the Natural Science Foundation of Hunan Province (2025JJ50493), and Health Research Project of Hunan Provincial Health Commission (W20243173).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare no competing interests. Generative AI and AI-assisted technologies were NOT used in the preparation of this work.