Lactylation in cancer: molecular mechanisms and advances in clinical study.

Introduction
Lactate has long been considered a metabolic byproduct of glycolysis, especially known for its accumulation in the tumor microenvironment (TME) due to the “Warburg effect.” However, this conventional view has been fundamentally challenged in recent years. In 2019, Zhao et al. reported a novel post-translational modification of lysine, known as Kla, marking a pivotal shift in understanding lactate—from a mere metabolic waste to a critical signaling molecule [1]. As an emerging epigenetic modification, lactylation is widely present across various tissues and contributes to gene expression regulation by modifying histones. In the context of cancer, particularly within the TME, the high levels of aerobic glycolysis in cancer cells lead to excessive lactate production, providing a substrate foundation for lactylation. Studies have shown that lactylation plays a critical role in regulating tumor cell proliferation, immune evasion, stemness maintenance, and drug resistance [2]. In addition, an increasing number of studies have focused on lactylation as a potential therapeutic target and biomarker. Inhibition of lactylation has been shown to restore T cell function and enhance the efficacy of immunotherapy [3], while modulation of lactylation levels can influence the sensitivity of cancer cells to chemotherapy, offering new strategies to overcome drug resistance [4, 5]. In this review, we summarize current knowledge on lactylation from multiple perspectives, including its molecular mechanisms, regulation within the TME, organ-specific effects across cancers, translational and clinical progress, and future research directions. By integrating findings from basic research with emerging biomarker and therapeutic evidence, we aim to clarify how lactylation participates in tumor initiation and progression and to outline its potential as a target for precision cancer therapy. Although most mechanistic studies of lactylation have so far been conducted in cellular and animal models, a growing body of patient-derived data and early clinical studies has begun to underscore its translational relevance. Correlative analyses already link elevated histone lactylation in human tumors to aggressive clinicopathological features and poor prognosis, and several clinical trials targeting lactate metabolism have demonstrated that systemic lactate levels can be modulated in patients. Thus, while lactylation currently remains largely a preclinical concept, it is rapidly evolving into a field of growing biomarker and therapeutic interest that is only just beginning to enter the translational and clinical arena.

Overview of Post-Translational Modifications (PTMs)
Post-translational modifications (PTMs) regulate protein function and cellular activity by adding or removing chemical groups after protein synthesis, thereby altering protein conformation, stability, localization, and interactions [6, 7]. Because PTMs coordinate key processes such as signal transduction and gene regulation, their dysregulation is widely implicated in cancer [8]. Accordingly, canonical PTMs including phosphorylation, acetylation, methylation, and ubiquitination have been extensively studied as drivers of tumor progression and therapy resistance and as therapeutic targets [9–13]. More recently, Kla has emerged as a metabolically linked PTM in cancer [14] (Fig. 1). Lactylation involves the covalent attachment of a lactate molecule to lysine residues, a process influenced by the metabolic state of the cell. This modification has been shown to regulate several processes that are dysregulated in cancer, including metabolic reprogramming, gene expression, and immune cell function [15]. As an example, lactylation is involved in the Warburg effect, a hallmark of many cancers, where cells predominantly rely on glycolysis and lactate production for energy, even in the presence of oxygen. However, the precise molecular mechanisms and crosstalk with other PTMs remain poorly understood, warranting further investigation. Notably, several lysine residues can host multiple acyl marks (e.g., histone H3K18/K23), and shared ‘writer’/‘eraser’ enzymes (p300/CBP; HDAC1–3 and SIRT1–3) act on both lactylation and acetylation. This overlap raises the possibility of competition or synergy that shapes chromatin accessibility and protein function [16]. Understanding how lactylation interacts with other PTMs and contributes to cancer progression is crucial for identifying novel therapeutic strategies that target metabolic reprogramming in tumors [17].

Emergence of lactylation as a novel PTM
Lactylation is a newly identified post-translational modification in which lactate is covalently conjugated to lysine residues, thereby directly coupling cellular metabolic status to protein function [18]. Because this modification reflects intracellular energy flux, it is particularly relevant to cancers characterized by metabolic rewiring [19]. In this context, the Warburg effect—whereby tumor cells preferentially rely on glycolysis for ATP generation even under aerobic conditions—drives sustained lactate accumulation [20], providing a permissive metabolic milieu for lactylation and enabling lactate to function as a bioactive metabolite rather than a mere waste product. Functionally, lactylation has been linked to the regulation of transcriptional programs, including modulation of hypoxia-inducible factor 1α (HIF-1α) activity, which supports angiogenesis, survival, and metastatic potential under hypoxic stress [21, 22]. Beyond tumor-intrinsic effects, lactylation also shapes anti-tumor immunity; accumulating evidence suggests that it promotes tumor-associated macrophage polarization toward an M2-like immunosuppressive phenotype, thereby reinforcing a tumor-permissive microenvironment [23, 24]. Together, lactylation constitutes a metabolism-coupled regulatory axis integrating tumor metabolic adaptation with immune remodeling during cancer progression [25].

Importance of lactylation in tumor progression and therapy resistance
Lactylation plays a significant role in cancer progression by integrating metabolic changes with gene expression regulation, immune evasion, and therapeutic resistance. While its discovery as a PTM initially linked it to metabolic reprogramming, growing evidence suggests that lactylation influences broader tumorigenic processes beyond metabolism. Cancer cells, particularly those experiencing hypoxia or increased glycolytic activity, accumulate lactate, which fuels histone and non-histone lactylation to activate pro-tumorigenic transcriptional programs [26, 27]. This regulation extends beyond metabolic enzymes and affects pathways related to angiogenesis, cell proliferation, and epithelial-mesenchymal transition (EMT), all of which contribute to tumor progression [28, 29]. Unlike acetylation, which generally enhances gene expression through histone modifications, lactylation exhibits context-dependent effects, selectively activating genes that support tumor cell survival under metabolic stress [30]. As such, lactylation represents a critical link between the metabolic state of a tumor and its ability to adapt to environmental challenges.
Beyond its intrinsic effects on cancer cells, lactylation has emerged as a key modulator of the TME [31]. As discussed in the previous section, lactylation influences macrophage polarization, skewing them toward an M2-like, pro-tumorigenic phenotype. However, its role in immune regulation extends beyond macrophages. Recent studies indicate that lactylation affects the function of other immune cells, including T cells and dendritic cells, potentially suppressing anti-tumor immunity and fostering an immune-privileged tumor niche [32, 33]. Additionally, lactylation-mediated metabolic reprogramming in cancer-associated fibroblasts (CAFs) enhances their ability to secrete extracellular matrix components and growth factors, further promoting tumor invasion and resistance to therapy [34]. These findings underscore lactylation’s role not only in individual cancer cells but also in the broader tumor ecosystem, where it contributes to a permissive and immune-suppressive microenvironment.
Lactylation also contributes to therapy resistance through multiple mechanisms (Fig. 2). First, by epigenetically activating stress-response and survival genes, lactylation enables tumor cells to withstand cytotoxic agents, including chemotherapy and radiation [15]. For example, recent work by Chen and colleagues provided a compelling illustration of how non-histone lactylation can be directly linked to therapy resistance in clinically relevant models. They showed that lactylation of the DNA damage response factor NBS1 enhances homologous recombination repair and promotes resistance to platinum-based chemotherapy, whereas pharmacologic inhibition of NBS1 lactylation using stiripentol, an antiepileptic drug already approved for Dravet syndrome, restored chemosensitivity in multiple tumor models. Building on these preclinical data, the same group has initiated a prospective clinical trial testing stiripentol in combination with chemotherapy and immunotherapy for patients with peritoneal metastatic carcinoma (ChiCTR2400083649), representing one of the first rational attempts to repurpose an approved drug as an anti-lactylation agent in the clinic [35]. In addition, lactylation has been implicated in the upregulation of ABC transporters, which enhance drug efflux and reduce intracellular drug accumulation, a common mechanism underlying multidrug resistance [36]. Second, lactylation modulates metabolic enzymes to favor adaptive energy production pathways, allowing cancer cells to evade apoptosis even under therapeutic stress [29]. Moreover, the synergistic or competitive relationships between lactylation and other PTMs (e.g., acetylation and ubiquitination) may further regulate tumor phenotypes and therapeutic responses [37], we will discuss this in detail below. Finally, lactylation-driven immune suppression within the TME reduces the efficacy of immunotherapies, such as immune checkpoint inhibitors (ICIs), by dampening anti-tumor immune responses [38]. These findings suggest that targeting lactylation may enhance the efficacy of current treatments by reversing drug resistance and restoring immune sensitivity.
Given its role in tumor progression and therapy resistance, lactylation is an attractive target for cancer therapy. Although research on lactylation inhibitors is still in its early stages, potential therapeutic strategies include targeting lactate metabolism, inhibiting lactylation-associated enzymes, or developing combination therapies that enhance existing treatment responses [39]. Future studies are needed to explore the specificity and reversibility of lactylation, as well as its broader implications across different cancer types. As our understanding of lactylation deepens, targeting this PTM could pave the way for novel treatment strategies aimed at disrupting tumor metabolism, overcoming drug resistance, and restoring immune function in cancer patients.

Mechanisms of lactylation in cancer

Enzymes and pathways involved in lactylation
The mechanisms of lactylation can be broadly categorized into enzymatic and non-enzymatic pathways. The two chiral isomers of lactate, L-lactate and D-lactate, are involved in distinct lactylation pathways, with L-lactate primarily contributing to enzymatic lactylation and D-lactate associated with non-enzymatic modifications (Fig. 3) [40].
Current literature supports two biochemically distinct enzymatic lactylation routes that are defined by the chemical nature of the activated donor used to transfer the L-lactyl group. The first route employs L-lactyl-CoA, whereas the second relies on an ATP-derived L-lactyl-AMP intermediate. Both processes selectively generate the L-stereoisomer and together explain the bulk of regulated Kla observed in eukaryotic cells.
Recent studies show that enzymatic lactylation supplied by L-lactate proceeds along two chemically independent routes that converge on lysine side chains. In the L-lactyl-CoA route, nuclear phosphorylated acetyl-CoA synthetase 2 (ACSS2) acts as a lactyl-CoA synthetase (L-lactate + CoA + ATP → L-lactyl-CoA + AMP + PPi), couples physically with the GCN5-related N-acetyltransferase (GNAT) acetyl-transferase lysine acetyltransferase 2 A (KAT2A) and feeds a high-energy thio-ester directly into the catalytic pocket, driving installation of H3K18la and H3K23la and boosting Wnt and PD-L1 transcription to promote immune evasion [41]. The same thio-ester is accepted by the co-activator p300 and its paralogue CREB-binding protein (CBP), which write H4K12la at active promoters and enhancers in diverse cancers and stem cells, and by the (MOZ-Ybf2/Sas3-Sas2-Tip60) MYST enzyme histone-acetyltransferase-bound-to-ORC1 (HBO1), whose preference for H3K9la stimulates invasion programmes in solid tumors [42, 43]. Collectively, these findings position lactyl-CoA dependent lysine acetyltransferases (KATs) as the principal chromatin writers that translate glycolytic flux into gene expression. A second enzymatic solution uses an ATP-derived acyl-adenylate. Cytosolic alanyl-tRNA synthetase 1 (AARS1) and mitochondrial AARS2 bind L-lactate together with ATP to form an L-lactyl-AMP intermediate that is transferred directly to lysine residues on non-histone substrates. AARS1 lactylates Yes-associated-protein (YAP) at K90 and TEA-domain-transcription-factor-1 (TEAD1) at K108, forming a positive feedback loop that strengthens Hippo pathway output in gastric cancer [44], whereas AARS2 modifies cyclic-GMP-AMP synthase (cGAS), pyruvate-dehydrogenase-E1-alpha (PDHA1) and carnitine-palmitoyltransferase-2 (CPT2), dampening DNA sensing and oxidative metabolism when mitochondrial lactate accumulates [45]. These observations confirm that aminoacyl-tRNA synthetases function as metabolite sensors that broadcast lactate availability beyond the nucleus. Non-enzymatic lactylation, in contrast, originates from the glycolytic by-product methylglyoxal. Detoxification byglyoxalase-1 (GLO1) produces S-lactoyl-glutathione, which is normally hydrolysed by GLO2. Immune activation suppresses GLO2, causing cytosolic build-up of S-lactoyl-glutathione that transfers an R-configured D-lactyl group to nearby cysteine and then to lysine. Proteomic profiling of activated macrophages identified more than two thousand K(D-la) sites, and modification of RelA at K310 markedly reduces NF-κB transcriptional activity, creating a negative feedback brake on inflammation [46].
Together, these data establish L-lactyl-CoA and L-lactyl-AMP as the two enzymatic donors that install stereospecific L-lactyl marks on histone and non-histone proteins, while a chemically distinct D-lactyl pathway arises from spontaneous transfer of the glycolytic intermediate S-lactoyl-glutathione, highlighting how cellular metabolism routes carbon overflow into tailored post-translational signaling.

Structural insights
To gain a deeper understanding of the molecular mechanisms and biological significance of lactylation, we provide here a comparative overview of the structures of known “writer” enzymes (which add lactyl groups) and “eraser” enzymes (which remove lactyl groups) involved in lactylation. Accumulating evidence indicates that several acetyltransferases act as genuine lysine lactyltransferases. These include the KAT3 co-activators p300 and CBP, the MYST family members KAT5/TIP60 and HBO1/KAT7, and the non-canonical alanyl-tRNA synthetases AARS1 and AARS2. In contrast, zinc-dependent class I histone deacetylases HDAC1–3 and the NAD⁺-dependent sirtuins SIRT1–3 serve as the main lysine delactylases in cells, with SIRT3 standing out as a crucial mitochondrial eraser that also targets a subset of nuclear proteins. Focusing on the structures of these enzymes, in particular their acyl-binding pockets, catalytic motifs and subcellular distribution, helps to clarify why they prefer L-lactyl donors and how they generate the diverse functional consequences of lactylation. By examining the structural differences among p300/CBP, AARS1/2, HDAC1–3, and SIRT1–3, we highlight how their architectures underlie their functions and substrate specificities – including stereoselectivity for the L-lactyl modification [29] (Table 1) (Fig. 4).

p300/CBP lysine lactyltransferases
The histone acetyltransferases (HATs) p300 and CBP (KAT3B/KAT3A) are large multidomain enzymes whose catalytic HAT domains can promiscuously accommodate various acyl-CoA substrates [47]. A high-resolution crystal structure of the p300 HAT domain bound to a Lys-CoA analog revealed a shallow, flexible substrate pocket that explains its broad acyl-donor specificity. Unlike Gcn5/PCAF acetyltransferases, p300/CBP lack a strict catalytic base; instead, they appear to employ a “hit-and-run” (Theorell–Chance) mechanism in which acetyl (or lactyl) transfer occurs without a stable enzyme–substrate ternary complex [48]. Key active-site residues (e.g. a Trp-loop and an autoinhibitory loop that must rearrange for catalysis) create an open cleft accommodating larger acyl groups [49–51]. This structural plasticity enables p300/CBP to utilize L-lactyl-CoA as a cofactor for Kla, as demonstrated in vitro. Consistently, p300-catalyzed histone lactylation is observed only with the natural L-lactate isomer, whereas the D-lactyl modification arises from non-enzymatic glyoxalate chemistry. These findings support a functional preference for the L-lactyl donor in enzymatic reactions, although explicit stereoselectivity against D-lactyl-CoA remains to be fully established [52]. Beyond histones, CBP/p300 can also lactylate non-histone targets. For example, the CBP HAT was recently shown to modify the DNA-repair factor MRE11 at K673 with a lactyl group in response to DNA damage. Notably, that study found CBP’s phosphorylation by ATM enhanced this activity, highlighting that the same structural features (an accessible acyl-CoA binding cleft and lysine-interacting groove) enable CBP to act on diverse protein substrates when appropriately activated. In sum, the p300/CBP HAT domain’s structural flexibility underlies its ability to catalyze Kla. Broad acyl-pocket adaptations – which allow favorable binding of the lactyl moiety’s extra hydroxyl-bearing carbon – and the lack of a stringent catalytic base both contribute to efficient transfer of the larger lactyl group [53]. Collectively, the structural plasticity and autoinhibitory-loop control of the p300/CBP HAT core explain their capacity to catalyze L-lactylation in defined contexts; however, current evidence indicates that HBO1/KAT7 also serves as a major histone lactyltransferase, underscoring enzyme- and context-specific contributions to cellular lactylation [43].

AARS1/2 alanyl-tRNA synthetases as lactyltransferases
AARS1 and AARS2 (cytosolic and mitochondrial alanyl-tRNA synthetases (AlaRS)) have emerged as unusual lactylation “writers” that directly attach lactate onto lysine side chains [54, 55]. Canonically, AlaRS catalyzes ATP-dependent alanine activation (Ala-AMP) followed by transfer to Ala 3’ end. Structurally, AARS1/2 share a class II aminoacyl-synthetase fold with an alanine-binding pocket shaped by key specificity residues (AARS1: M46, R77, N216, D239, G241; AARS2: M79, R110, N242, D265, G267) that coordinate the substrate carboxylate and enforce selectivity [56]. Owing to the close structural similarity between alanine and lactate (α-hydroxypropionate), the same pocket can accommodate L-lactate and support formation of a lactyl-AMP intermediate [54]. In this moonlighting reaction, AARS1/2 adenylate L-lactate and then transfer the lactyl group to the ε-amino group of lysine on protein substrates. Docking and mutational analyses support lactate engagement by alanine-recognition residues, although a lactate-bound co-crystal structure is still lacking [44, 45]. AARS1/2 preferentially utilize the L-lactate enantiomer, consistent with the chiral architecture of the active site and the cellular predominance of L-lactate under glycolytic conditions [45, 57]. In addition, the canonical C-terminal editing module that proofreads mischarged tRNA Ala is likely bypassed in Kla because lysine side chains serve as acceptors rather than tRNA substrates. Clinically relevant “hyper-lactylation” mutations in AARS2 cluster near the substrate pocket; modeling of variants such as R199C suggests subtle reshaping/charge changes that enhance lactyl-AMP generation and protein lactylation [58]. Collectively, AARS1/2 illustrate a distinctive structure–function adaptation in which an aminoacylation active site is repurposed to link intracellular lactate to lysine lactylation.

HDAC1–3 zinc-dependent delactylases
The class I histone deacetylases HDAC1, 2, and 3 have been identified as major “erasers” of Kla on histones [59]. They share a conserved Zn2⁺-dependent catalytic core with a tubular active site, where Zn2⁺ is coordinated by one histidine and two aspartates; structural studies (e.g., HDAC8) show a narrow tunnel that positions the acyl-lysine amide bond adjacent to a Zn2⁺-activated water molecule [60]. The same hydrolytic mechanism used for deacetylation applies to delactylation: zinc-bound water attacks the acyl carbonyl to form a tetrahedral intermediate, and a nearby conserved tyrosine protonates the leaving amine to release free lysine and acetate/lactate [61, 62]. Modeling indicates the pocket can accommodate ε-N-lactyllysine (slightly larger than acetyl and bearing an α-hydroxyl) without major rearrangements [59, 63]. In vitro, HDAC1–3 efficiently remove ε-L-lactyl-lysine, with HDAC3 particularly potent on histone peptides; SIRT2 also delactylates histones but generally with lower catalytic efficiency in comparable peptide assays. Consistent with an active site that primarily recognizes the amide linkage, zinc-dependent HDACs show no strong reported preference for L- versus D-lactyl-lysine (unlike sirtuins) [59, 64, 65]. In cells, HDAC1 and HDAC3 appear to dominate histone delactylation, as sirtuin inhibition (nicotinamide) has little impact on global histone lactylation, supporting a key role for nuclear HDAC1/2/3 in removing lactyl marks on chromatin [61, 62].

SIRT1–3 NAD+ dependent delactylases
Sirtuins 1–3 (class III HDACs) are NAD⁺-dependent deacylases that also contribute to delactylation [66]. Structurally, SIRT1–3 share a conserved Rossmann-fold core that binds NAD⁺ and an acyl-lysine substrate, and catalysis proceeds via transfer of the lysine acyl group to NAD⁺ to form an O-acyl-ADP-ribose intermediate, ultimately yielding deacyl-lysine, nicotinamide, and an ADP-ribose product [67–69]. A conserved catalytic histidine (e.g., SIRT1 H363; SIRT2 H187) and an acyl pocket adjacent to the NAD⁺ site underlie activity toward diverse acyl groups; notably, the hydrophobic pocket is relatively flexible in SIRT2, supporting tolerance of bulkier modifications [70]. Consistent with this architecture, structural and biochemical studies identify SIRT2 as an efficient lysine delactylase capable of removing ε-lactyl-lysine from histone peptides/nucleosomes and reducing histone Kla in cells, although both efficiency and stereochemical bias can be substrate- and condition-dependent [64, 71]. SIRT3 has a comparatively narrower acyl pocket optimized for shorter acyl chains [72], yet it can also de-lactylate specific histone sites (e.g., H3K9 and H4K16) in defined contexts [65] and may act on nuclear substrates under stress, including lactylated proteins such as CCNE2 [73]. SIRT1, the major nuclear sirtuin, shares the same catalytic core and has been shown in vitro to remove lactyl groups from histone peptides alongside SIRT2/3 [74]. Overall, SIRT1–3 function as NAD⁺-dependent delactylases with isozyme-specific pocket features that shape substrate scope; consistent with compartmental roles, SIRT2 (cytosolic/nuclear) and SIRT3 (mitochondrial with stress-linked nuclear activity) target both histone and non-histone lactylation [71, 74, 75]. Together with HDAC1–3, which dominate chromatin delactylation in cells, sirtuins help maintain lactylation dynamics and substrate selectivity [59].

“reader” proteins
Beyond the well-characterized “writer” and “eraser” enzymes, recent studies have revealed that “reader” proteins also play important roles in Kla. These factors do not catalyze the addition or removal of the modification; instead, they specifically recognize lactyl-lysine marks and translate this metabolic signal into distinct transcriptional or chromatin-remodeling outcomes—a concept consistent with how chromatin readers decode other histone PTMs to recruit regulatory complexes [76]. Emerging primary evidence identifies several Kla readers: DPF2 binds H3K14la (shown by photoaffinity-probe proteomics, biochemistry, and CUT&Tag) and couples lactylation to oncogenic transcription and cancer-cell survival, establishing DPF2 as a bona fide Kla effector [77]. During early iPSC reprogramming, BRG1/SMARCA4 is selectively recruited to H3K18la-enriched promoters of pluripotency and epithelial-junction genes, acting as a chromatin-remodeling reader that links the metabolism–lactylation axis to gene activation [78]. In immune regulation, a systematic screen of 28 bromodomains found that the TRIM33 bromodomain uniquely recognizes histone Kla peptides, nominating TRIM33 as a Kla-selective reader connected to late-stage macrophage transcriptional programs [79]. Although the current roster remains limited, these studies demonstrate that lactylation is an information-rich chromatin signal that can be decoded by specific reader modules; nevertheless, more systematic and in-depth structural and functional analyses of Kla readers are still needed [80].

Molecular targets of lactylation in cancer cells
Kla was initially discovered on histones, and chromatin remains the most extensively characterized substrate class [81]. Multi-omics profiling of patient-derived tumors, such as pancreatic and colorectal cancers, has identified prominent histone lactylation marks, including H3K18la and H4K12la, as key lactylation sites. Among these, H3K18la has been particularly highlighted due to its critical roles in reshaping the tumor immune microenvironment, regulating tumor progression, and modulating tumor-associated macrophage functions [82]. Genome-wide mapping further demonstrates a pronounced accumulation of H3K18la at the promoters of VEGFA, HK2 and SOX2; removal of this mark by CRISPR substitution of H3K18 with arginine ablates the associated transcriptional programmes and slows xenograft growth, confirming that histone lactylation functions as an activating epigenetic signal [83]. As mentioned above, beyond the well-known writer p300, the acetyl-transferase HBO1/KAT7 has been identified as a dedicated histone lactyltransferase: HBO1 preferentially installs H3K9la, sustains MYC- and WNT-responsive gene sets, and its deletion curtails tumor expansion in triple-negative breast-cancer models [43].
Proteome-scale “lactylome” studies reveal that lactylation also decorates more than a thousand non-histone proteins that govern metabolism, DNA repair and oncogenic signaling. In hepatocellular carcinoma, lysine-28 lactylation of adenylate kinase-2 impairs ATP buffering and drives proliferation and pulmonary metastasis [19]. In glioblastoma, glucose-driven metabolic reprogramming induces lactylation of the DNA-repair scaffold XRCC1, strengthening its interaction with PARP1 and conferring resistance to temozolomide and radiotherapy [84]. Core glycolytic enzymes are frequent targets: phosphoglycerate-kinase-1 and enolase-1 acquire lactyl marks that enhance catalytic turnover and bolster intracellular lactate production, whereas lactylation of PDHA1 inhibits pyruvate entry into the tricarboxylic-acid cycle, locking tumor cells into aerobic glycolysis [85]. Lactylation also tunes oncogenic signal transduction. Modification of AKT1 at lysine 14 prolongs its phosphorylated state, amplifying PI3K–AKT signaling, while lactylated STAT3 at lysine 685 exhibits enhanced nuclear retention and transcriptional output, collectively promoting cell-cycle progression and immune evasion. Chromatin regulators add an additional layer of control: lactylation of DNMT1 weakens maintenance DNA methylation, whereas lactylated EZH2 disrupts PRC2 assembly, together creating a more permissive chromatin landscape for oncogene expression [29]. Beyond tumor-intrinsic roles, lactylation reshapes the immune micro-environment. In tumor-associated macrophages (TAMs), extracellular lactate drives genome-wide H3K18la that activates ARG1 and MRC1, promoting M2 polarization and dampening anti-tumor immunity; inhibition of lactate transport reverses these marks and restores macrophage cytotoxicity [86]. Similar lactyl-driven immunomodulatory patterns are emerging in dendritic and T cells, suggesting a broader role for the modification in sculpting an immune-suppressive niche. In the following sections, we will focus on the mechanistic studies of lactylation in specific types of cancer, systematically exploring its functional roles across different tumor categories. Through these investigations, we aim to identify potential therapeutic targets and provide new directions and theoretical foundations for future precision cancer therapy.

Impact of lactylation on tumor metabolism
Lactate is the defining metabolite of the Warburg phenotype in cancer cells, and the discovery of Kla has revealed a direct molecular bridge between lactate accumulation and the reprogramming of gene and protein function [87]. At the chromatin level, histone marks such as H3K18la and H3K23la accumulate at promoters of key metabolic genes, relax nucleosomal architecture, and enhance transcription across several tumor types [1, 88]. Beyond chromatin, lactylation inactivates mitochondrial enzymes including PDHA1 at Lys336 and CPT2 at Lys457 and Lys458, redirecting carbon flux toward glycolysis [55]. Fatty-acid synthase and other biosynthetic enzymes also undergo lactylation, indicating that lipid anabolism is similarly regulated [19, 89]. By coordinately modulating glucose, glutamine, lipid, and oxidative phosphorylation pathways, lactylation furnishes cancer cells with a rapid and adaptable metabolic programme that supports macromolecular synthesis, maintains an acidic and immunosuppressive microenvironment, and ultimately accelerates tumor progression and therapeutic resistance.

Promoting aerobic glycolysis (Warburg effect)
Cancer-associated lactate accumulation drives Kla of both histone and metabolic proteins, creating a feed-forward circuit that sustains aerobic glycolysis (Warburg effect) [90]. Notably, histone H3 lactylation at K18 (and to a lesser extent K23) is enriched at the promoters of key glycolytic enzyme genes (e.g. HK2, PFKFB3, LDHA) under high-lactate conditions [91, 92]. This epigenetic mark is associated with increased chromatin accessibility and RNA Polymerase II occupancy, thereby activating transcription of these glycolytic genes [93, 94]. In pancreatic ductal adenocarcinoma, lactate accumulation fuels the enrichment of histone H3 lysine-18 lactylation (H3K18la) at the promoters of TTK and BUB1B, generating a glycolysis–H3K18la–TTK/BUB1B positive-feedback circuit that persistently augments glucose uptake and lactate production, thereby reinforcing aerobic glycolysis [90]. Under hypoxic conditions, the mitochondrial Ala-tRNA synthetase AARS2 catalyzes lactylation of PDHA1 at lysine 336, blocking pyruvate entry into oxidative phosphorylation and redirecting carbon flux toward glycolysis, thereby accelerating the Warburg metabolic program [55]. In a FOLFOX-resistant hepatocellular carcinoma model, elevated H3K14la upregulates NEDD4, triggering PTEN degradation and activating PI3K–Akt signaling; this cascade further increases HK2, PFKFB3, and LDHA expression, establishing a self-reinforcing loop that drives high glycolytic flux [95]. In gastric cancer, high expression of the glucose transporter GLUT3 enhances lactate production by upregulating LDHA, elevates global histone lactylation levels, and the extent of lactylation positively correlates with the cells’ highly glycolytic phenotype [96]. In VHL-deficient clear cell renal cell carcinoma, H3K18la accumulates at the PDGFRβ promoter to activate its transcription; the ensuing PDGFRβ signalling, in turn, further boosts lactylation, thereby sustaining the abnormally heightened glycolytic flux [97]. Taken together, lactylation is not only a by-product of lactate metabolism but also establishes metabolic-epigenetic positive-feedback loops that broadly propel aerobic glycolysis in tumor cells, thereby acting as a pivotal sustaining mechanism of the Warburg phenotype [85].

Enhanced lactate efflux and maintenance of an acidic microenvironment
Extensive aerobic glycolysis raises intracellular lactate concentrations, and cancer cells avert lethal acidification by over-expressing proton-linked monocarboxylate transporters, particularly MCT4 (SLC16A3), which efficiently co-exports lactate and H⁺ ions [98]. Histone H3 lysine 18 lactylation becomes enriched at the promoter of HIF1A under sustained lactate exposure, sustaining HIF-1α transcriptionand may thereby promote MCT4 expression through the established HIF-1α-dependent pathway, potentially forming a positive feedback loop between lactate production and efflux that still awaits experimental confirmation [99, 100]. Up-regulated MCT4 assembles with its chaperone CD147 at invadopodia, where the cotransport of lactate and protons not only normalises cytosolic pH but also acidifies the pericellular milieu [101]. The resulting extracellular pH can drop below 6.8, a range that activates cathepsins and matrix metalloproteinases, thereby accelerating extracellular-matrix degradation and facilitating invasion [102, 103].

Regulation of glutamine metabolism (glutaminolysis)
Glutamine is the most abundant plasma amino acid and, in many tumors, becomes an indispensable nutrient: its carbon feeds the tricarboxylic-acid cycle as α-ketoglutarate, its nitrogen donates to nucleotide and hexosamine synthesis, and its catabolism through malic enzyme yields NADPH for redox control and lipid biosynthesis [104]. Cancer cells import this fuel chiefly via the high-affinity transporter ASCT2 (SLC1A5) [105]. Once inside the mitochondria, glutaminase (GLS1) converts glutamine to glutamate, which is de-aminated to α-ketoglutarate, replenishing TCA intermediates and sustaining rapid ATP and biosynthetic precursor production; oncogenic c-Myc amplifies this pathway by transcriptionally up-regulating both ASCT2 and GLS1, driving the well-known “glutamine addiction” phenotype of proliferating cancer cells [106]. Recent studies indicate that elevated lactate triggers p300-mediated H3K18 Kla to accumulate at the c-Myc promoter, thereby enhancing both the transcription and stability of c-Myc [107]; c-Myc subsequently transcriptionally activates the glutamine transporter SLC1A5/ASCT2 and the glutaminase GLS1, markedly enhancing glutamine uptake and catabolism and establishing the epigenetic groundwork for a “lactate–c-Myc–glutaminolysis” positive feedback loop; however, this loop still awaits systematic experimental confirmation [108]. At the non-histone level, lactate promotes lactylation of NMNAT1 at Lys128, augmenting its nuclear localization and enzymatic activity; this sustains the NAD⁺ salvage pathway and enables pancreatic cancer cells to survive glucose limitation by relying on glutaminolysis [105]. Meanwhile, tumor cells express a mitochondria-targeted variant of SLC1A5 that channels glutamine directly into the organelle, fuelling TCA-cycle anaplerosis and NADPH production; in a high-lactate, high-lactylation milieu, this mitochondrial supply cooperates with nuclear-cytoplasmic regulatory circuits to further entrench glutamine “addiction” [109]. In sum, lactylation integrates glutaminolysis through a triple-layered strategy—epigenetic activation, non-histone modification, and subcellular transport—establishing a pivotal metabolic nexus that sustains tumor growth and survival under nutrient stress.

Regulates lipid synthesis and fatty acid metabolism.
Cancer cells rewire lipid metabolism on two fronts: they boost de novo fatty-acid synthesis by up-regulating ACLY, ACC, and especially fatty-acid synthase (FASN), so that even in lipid-rich environments endogenous palmitate production can furnish membranes, signalling lipids, and an NADPH “sink” that buffers oxidative stress [110, 111]; in parallel, invasive or therapy-resistant subpopulations switch on fatty-acid uptake and β-oxidation, importing long-chain acyls via CD36/FATP transporters and routing them through CPT1-dependent oxidation to generate ATP and acetyl-CoA when glucose is scarce, a dependency that sustains metastatic seeding and can be blunted by pharmacological FAO blockers [112, 113]. Together, the anabolic (lipogenesis) and catabolic (FAO) arms of fatty-acid metabolism supply bio-building blocks, energy, and redox currency, making enzymes such as FASN and CPT1 attractive, clinically tractable targets across multiple tumor types. Accumulating evidence indicates that lactylation plays a critical role in regulating lipid metabolism in cancer cells. In hepatocellular carcinoma, lactate-induced histone lactylation, especially H3K18la, promotes lipid metabolic reprogramming by enhancing YTHDC1 transcription. YTHDC1 stabilizes the m6A-modified long non-coding RNA NEAT1, which recruits p300 to the promoter of stearoyl-CoA desaturase (SCD), increasing H3K27 acetylation and activating SCD expression. SCD catalyzes the conversion of saturated to monounsaturated fatty acids, supporting membrane synthesis and tumor cell proliferation. This pathway links lactate-driven epigenetic modification to enhanced lipid synthesis and fatty acid desaturation in cancer [114]. A recent study demonstrated that lactate-induced histone H3K18la upregulates the expression of ACAT2, which catalyzes the conversion of acetyl-CoA to acetoacetyl-CoA, thereby promoting the biosynthesis of cholesterol precursors. Notably, ACAT2 significantly facilitates the transport of cholesterol to TAM via small extracellular vesicles (sEVs). These sEVs, enriched in cholesterol, are internalized by macrophages and induce their polarization toward an immunosuppressive M2 phenotype, contributing to the establishment of an immunosuppressive TME. sEVs derived from ACAT2-overexpressing tumor cells exhibit elevated cholesterol levels and activate canonical M2-polarizing signaling pathways, including PI3K/Akt, JAK/STAT3, and STAT6, in recipient macrophages. Conversely, ACAT2 depletion reduces the cholesterol content in sEVs, thereby inhibiting M2 polarization and enhancing CD8⁺ T cell activity. Furthermore, pharmacological inhibition of sEV biogenesis with GW4869 or blockade of sEV uptake using chlorpromazine significantly attenuates cholesterol transfer and its associated immunosuppressive effects [115].

Enhanced metabolic plasticity and stress adaptation
Metabolic plasticity is the ability of cancer cells to dynamically rewire glycolysis, oxidative phosphorylation, glutaminolysis, and lipid metabolism to match fluctuating nutrient supply and biosynthetic demand [116, 117]. Stress adaptation encompasses signaling and homeostatic mechanisms that detect and mitigate metabolic, oxidative and proteotoxic stresses through HIF-1α activation under low-oxygen conditions and autophagy-mediated recycling during nutrient deprivation [118, 119]. Together, these processes sustain proliferation in nutrient-poor/hypoxic niches, enable metastasis, and promote therapy resistance by shifting metabolic flux and engaging pro-survival pathways [120]. Accumulating evidence indicates that lactylation can reinforce metabolic plasticity and stress tolerance in tumors and the TME. In prostate cancer, tumor-derived lactate drives histone lactylation in TAMs and reprograms macrophage function. PI3K pathway suppression reduces lactate output, lowers TAM histone lactylation, and enhances phagocytic anti-tumor activity, but this benefit can be offset by compensatory Wnt/β-catenin activation [121]. Mechanistically, β-catenin/TCF4 upregulates LDHA to restore glycolysis and lactate secretion that sustains TAM H3K18la; in PTEN/p53-deficient settings, Wnt/β-catenin activation during PI3K inhibitor resistance reinstates tumor lactate secretion and TAM H3K18la, which can be blocked by PORCN inhibition. Wnt/β-catenin also induces MCT1 (SLC16A1), further strengthening lactate transport and this feedback loop [122, 123].
In colorectal cancer stem cells (CCSCs), lactate accumulation induces p300-catalyzed, HDAC1-removable H4K12la, which transcriptionally upregulates GCLC to increase glutathione synthesis, suppress lipid peroxidation, and inhibit ferroptosis, thereby promoting oxaliplatin resistance. Inhibiting p300, LDHA, or GCLC re-sensitizes CCSCs to chemotherapy in vitro and in vivo, supporting a lactate–lactylation–GCLC axis as a stress-adaptive mechanism [42]. In glioblastoma, Wang et al. reported that glioma stem cells (GSCs) show high glycolysis/lactate and elevated histone lactylation that supports proliferation under stress. CBX3 facilitates EP300 utilization of lactyl-CoA, and the lactate–CBX3–EP300 axis promotes immune-evasive programs (e.g., CD47 upregulation and STAT3 activation), enhancing GSC survival and limiting phagocytosis under hypoxic/inflammatory conditions; reducing lactate production with DCA decreases lactylation and impairs GSC adaptability [124].
A direct mechanistic link between nutrient stress, lactate, lactylation, and autophagy has also been delineated. Under serum/amino-acid deprivation, ULK1 phosphorylates LDHA (Ser196) to increase lactate production, which supports KAT5/TIP60-mediated lactylation of PIK3C3/VPS34 (Lys356, Lys781). VPS34 lactylation strengthens interactions with BECN1, ATG14, and UVRAG, increases lipid kinase activity and PtdIns3P, and is required for efficient autophagosome formation, SQSTM1/p62 turnover, and endolysosomal degradation; disrupting this axis impairs autophagic flux in muscle and cancer cells [125]. Broader reviews further suggest bidirectional crosstalk between lactylation and autophagy: lactylation of DCBLD1 can stabilize PPP enzymes and DCBLD1 itself to support redox balance by limiting autophagic degradation of PPP components, whereas lactylation of VPS34 and TFEB enhances PtdIns3P production, lysosomal activity, and autophagic flux (including TFEB stabilization via protection from WWP2-mediated ubiquitination). Collectively, lactylation can either potentiate or restrain autophagy in a context-dependent manner, helping tumors tune catabolism and integrate metabolic reprogramming with immune evasion and therapy resistance [126, 127].
Beyond autophagy, accumulating evidence indicates that lactylation also shapes ferroptosis sensitivity and chemoresistance in cancer (Fig. 5). In CCSCs, p300-driven H4K12la at the GCLC locus promotes glutathione synthesis and ferroptosis suppression, conferring oxaliplatin resistance [42]. In KRASG12D-driven tumors, ACAT2 mediates lactylation of GCLM to enhance glutathione production and ferroptosis resistance; inhibiting ACAT2 reduces GCLM lactylation and overcomes resistance in vivo [128]. In lung adenocarcinoma, ferroptosis-inducing conditions trigger SUMO2 K11 lactylation, promoting degradation of ACSL4 and attenuating ferroptosis; a cell-penetrating peptide that blocks SUMO2-K11 lactylation restores ferroptosis and sensitizes tumors to cisplatin and chemo-immunotherapy in preclinical models [129]. More broadly, studies and reviews support that high lactate/protein lactylation can buffer oxidative damage and lipid peroxidation, reducing sensitivity to ferroptosis inducers and conventional therapies across multiple tumor contexts [130].
These data motivate combination strategies that target lactate supply, transport, or lactylation writing. LDH inhibitors and dichloroacetate can reduce lactate production [131, 132], while MCT1/MCT4 inhibitors reshape lactate gradients and acidity in the TME [133, 134]. p300/CBP inhibitors (e.g., A-485) [135] and clinical agents such as the bromodomain inhibitor CCS1477 may suppress lactylation at stress-response/drug-resistance loci [136]. Combining lactate/lactylation-targeted agents with ferroptosis inducers or chemotherapy could re-sensitize lactylation-high tumors by weakening autophagy-linked stress tolerance and antioxidant defenses, although current support remains largely preclinical [137].

Crosstalk between lactylation and other PTMs
Most proteins undergo multiple PTMs, which often interact with one another in a phenomenon known as PTM crosstalk. This interplay allows the integration of diverse cellular signals and can significantly influence a protein’s structure, function, and activity. Lactylation, characterized by the addition of a lactyl group to the ε-amino group of lysine residues, is structurally and functionally similar to other lysine-targeted modifications such as acetylation and ubiquitination [138]. Crosstalk between lactylation and other PTMs can be broadly classified into two categories: One category is competitive crosstalk, for instance, when macrophages are exposed to bacterial stimuli, histone acetylation levels progressively decline, whereas lactylation levels show a corresponding increase [1]. In hepatic stellate cells, exogenous lactate has been found to induce lactylation, simultaneously inhibiting acetylation, indicating a competitive relationship between these two modifications [91]. Rao et al. found that in lung adenocarcinoma (LUAD), lactylation of cGAS at lysine 21 (K21) promotes its interaction with the proteasomal subunit PSMA4, thereby facilitating cGAS degradation. Conversely, phosphorylation of PSMA4 at serine 188 (S188) disrupts its association with cGAS, helping to stabilize cGAS protein levels. Lactylation enhances the activity of PIK3CB through modification at lysine 415 (K415), which in turn suppresses ULK1-mediated phosphorylation of PSMA4 at S188. As a result, reduced PSMA4 phosphorylation favors cGAS degradation, ultimately impairing interferon production and supporting tumor growth [139]. The other mode is synergistic crosstalk, in which lactylation cooperates with other PTMs to enhance protein function or gene expression in a coordinated manner. Recent studies reveal a synergistic crosstalk between lactylation and ubiquitination in gallbladder cancer. Lactylation of YY1 at K183 enhances its transcriptional activation of FBXO33, an E3 ligase that promotes polyubiquitination of p53 at K291 and K292. This leads to p53 degradation and facilitates EMT and metastasis. Lactylation thus amplifies ubiquitination through transcriptional control, linking metabolic signals to protein degradation and tumor progression [140].

Detection methods for lactylation
Lactylation lies at the interface between metabolism and epigenetic regulation, so robust and reproducible detection is critical for interpreting its biological roles. At present, most studies still rely on antibody-based readouts such as western blotting, immunofluorescence and immunohistochemistry using pan-Kla or site-specific antibodies [141, 142]. To ensure specificity, rigorous validation is required, including lot-to-lot testing, peptide competition assays with lactylated versus non-modified peptides, and, where possible, comparison with genetic or pharmacological perturbation of lactylation “writers” and “erasers” [143, 144]. Without such controls, pan-Kla signals may be confounded by cross-reactivity with other acyl-lysine modifications or by batch variability between antibody lots [145, 146]. Mass spectrometry (MS) remains the gold standard for unambiguous site identification and quantification [147, 148]. Modern LC–MS/MS workflows can localise Kla sites at single-residue resolution, distinguish co-existing modifications on the same peptide [149], and, with appropriate labelling or label-free approaches, provide semi-quantitative estimates of site occupancy (stoichiometry) [150]. Enrichment with Kla-specific antibodies or chemical probes improves sensitivity, but also reinforces the need for orthogonal validation [151]. Ideally, MS-based mapping should accompany antibody-based assays when new lactylation sites, substrates or pathways are proposed [148, 152]. A further technical challenge is to distinguish L-lactylation from potential D-lactylation, since mammalian cells contain both L- and D-lactate pools. Most current antibodies and MS pipelines implicitly assume L-lactylation, yet do not routinely resolve stereochemistry. Future method development, including stereochemically defined synthetic standards, stereospecific antibodies or chromatographic separation of L- and D-lactylated peptides, will be important to dissect the biology of each enantiomeric form [153]. Finally, experimental design must take into account potential artefacts. Acute “lactate spikes” in cell culture (for example, high-dose sodium L-lactate, lactate esters or methyl-lactate) [151, 154] and broad-spectrum HDAC or sirtuin inhibitors can massively increase global acyl-lysine signals, including Kla, in ways that may not reflect physiological regulation [59, 74]. Such treatments are useful for probe development and pathway discovery, but they can also exaggerate lactylation levels and alter the balance between different acyl modifications. Careful dosing, time-course experiments, and inclusion of vehicle controls, genetic controls and, where possible, endogenous stress conditions (hypoxia, nutrient deprivation) are therefore essential to avoid over-interpreting artefactual changes in lactylation [55, 155]. Together, these considerations highlight that rigorous assay validation and thoughtful experimental design are prerequisites for generating reproducible and biologically meaningful data on lactylation.

Lactylation in TME
TME is a highly dynamic and complex ecosystem consisting of tumor cells, stromal cells, immune infiltrates, extracellular matrix (ECM), and vascular networks [156]. CAFs, the predominant stromal component, actively remodel the ECM and secrete pro-tumorigenic growth factors—such as VEGF, FGF, and TGF-β—thereby facilitating tumor proliferation, invasion, angiogenesis, and immune resistance [157]. The immune cell compartment—including regulatory T cells, myeloid-derived suppressor cells, and M2-polarized TAMs—establishes an immunosuppressive milieu via cytokine secretion [158]. Under physiological conditions, lactate levels in blood and normal tissues typically range between 1.5 and 3 mM [159]. In contrast, extracellular lactate levels in tumor cells can rise to approximately 40 mM [160]. Recent studies have shown that lactylation is extensively involved in the modulation of TME.

Influence of lactylation on immune cells in the TME
Lactylation drives the formation of an immunosuppressive TME across various cancers [161, 162]. In the TME, macrophages tend to acquire an M2-like program, and elevated histone lactylation in TAMs has been linked to this polarization and functional reprogramming [1, 163]. In hepatocellular carcinoma, tumor-derived lactate induces H3K18la in TAMs, which transcriptionally activates NUPR1 and upregulates immunosuppressive mediators (e.g., IL10, TGF-β, Arg1), thereby impairing CD8⁺ T-cell activation [164]. Lactylation also reinforces T-cell–mediated immunosuppression: Gu et al. reported that lactate promotes regulatory T-cell (Treg) differentiation and suppressive function by inducing MOESIN K72 lactylation, strengthening MOESIN–TGF-βRI interaction, activating SMAD3, and increasing FOXP3 expression [165]. Beyond these axes, a seminal study showed that lactate-driven H3K18la in tumor-infiltrating myeloid cells elevates METTL3, leading to m⁶A modification of Jak1 mRNA and activation of JAK1/STAT3 signaling; notably, METTL3 lactylation at K281/K345 enhances its RNA-binding affinity, and METTL3 blockade restores antitumor immunity by reducing M2-like macrophages/MDSCs and augmenting effector T-cell responses [166]. Consistent with these mechanistic insights, immunometabolism studies highlight that lactate/lactylation promote checkpoint programs (e.g., PD-L1, B7-H3) and immune dysfunction, facilitating Treg/TAM/MDSC accumulation and cytotoxic T-cell exhaustion [127, 167–169]. Conversely, lowering lactate levels or inhibiting lactate transport (MCT1/MCT4) can relieve immunosuppression and improve responses to immune checkpoint blockade in preclinical models and early translational studies, supporting rational combinations of lactate- or lactylation-targeting strategies with immunotherapy [170].

Lactylation and tumor stroma interaction
Lactate seeps into the surrounding stroma and stamps a chemical tag called lysine lactylation on stromal chromatin, rewiring how blood vessels, fibroblasts and immune barriers are built. In endothelial cells, VEGF-driven glycolysis raises the H3K9 lactyl mark on promoters of pro-angiogenic genes; wiping away the mark with HDAC2 or cutting glycolysis halts new-vessel sprouting, showing that H3K9la is an epigenetic regulator for tumor angiogenesis [171]. In fibroblast-like stroma, high lactate deposits H3K18la on the SOX9 promoter, turning quiescent fibroblasts into collagen-rich myofibroblasts that stiffen the matrix; LDHA knock-down lowers lactate, erases H3K18la and softens the desmoplastic rim [172]. CAFs send the signal back: in gastric cancer, CAF-secreted lysyl-oxidase boosts histone lactylation in neighbouring tumor cells, lifts PD-L1 expression and helps the tumor dodge immune attack [173]. In the TME, metabolic symbiosis is exemplified by the lactate shuttle, wherein tumor cells produce and export lactate as a result of aerobic glycolysis. This lactate is taken up by adjacent stromal cells, such as CAFs, via MCT1 transporters and is oxidized to generate ATP. This bidirectional metabolic coupling not only maintains stromal cell viability but also contributes to the reprogramming of the TME in favor of tumor growth. Targeting this lactate exchange axis is emerging as a promising therapeutic strategy to disrupt TME homeostasis and inhibit tumor progression [174].

Dual roles of lactate and lactylation in tumor immunity
Although lactate and protein lactylation have long been viewed mainly as pro-tumor and immunosuppressive regulators, recent work clearly supports a dual and context-dependent model. In tumor cells, lactate-driven lactylation frequently promotes DNA repair, survival and therapy resistance. Chen and colleagues showed that increased glycolytic flux and lactate accumulation in platinum-resistant gastric cancer are accompanied by upregulation of LDHA and elevated lactate levels in resistant tumors. They identified NBS1 lactylation at lysine 388 as a critical modification that stabilizes the MRE11–RAD50–NBS1 complex, enhances homologous recombination repair and protects tumor cells from DNA double-strand break–inducing therapies. TIP60 acted as the NBS1 lactyltransferase, whereas HDAC3 functioned as the delactylase; pharmacologic reduction of lactate production with the LDHA inhibitor stiripentol diminished NBS1 lactylation, impaired DNA repair and restored chemosensitivity in cell lines, patient-derived organoids and xenograft models [35]. These data exemplify a pro-survival axis in which tumor-intrinsic lactate and lactylation directly drive treatment resistance.
By contrast, lactate can exert anti-tumor effects when it primarily acts on the immune compartment. In a series of syngeneic mouse models, Feng et al. demonstrated that systemic administration of pH-neutral sodium lactate, but not glucose, suppressed MC38 colon carcinoma growth in a CD8⁺ T cell-dependent manner. Lactate treatment increased the proportion of stem-like TCF-1–expressing CD8⁺ T cells within tumors and enhanced their effector and memory potential, in part by inhibiting histone deacetylases and increasing H3K27 acetylation at the Tcf7 super-enhancer locus. CD8⁺ T cells preconditioned with lactate exhibited superior antitumor activity upon adoptive transfer, and lactate synergized with PD-1 blockade or a nanovaccine to produce durable tumor control [175]. Under these buffered conditions, lactate acts as a fuel and epigenetic modulator that supports, rather than suppresses, CD8⁺ T-cell-mediated immunity. The opposite effect emerges when lactic acid and glucose competition in the TME are considered. Liu et al. found that activated CD8⁺ T cells upregulate the high-affinity glucose transporter GLUT10, which is required for efficient glucose uptake, glycolysis and effector function in tumor-infiltrating lymphocytes. High concentrations of lactic acid in the TME bound an intracellular motif of GLUT10, reduced its glucose transport capacity, suppressed glycolysis and proliferation in CD8⁺ T cells and weakened their antitumor activity. Disrupting the interaction between lactate and GLUT10 with a mimetic peptide restored glucose use and synergized with GLUT1 inhibition or PD-1 blockade to improve tumor control [176]. This study illustrates how lactic acid can serve as an external brake on CD8⁺ T-cell metabolism and function, in clear contrast to the pro-immunogenic effects of systemic, pH-neutral lactate described above.
Recent reviews have therefore started to frame lactate and lactylation as dual regulators of tumor immunity and cancer biology. Hao et al. summaries how lactate accumulation and lactylation can suppress cytotoxic T cells and favor regulatory T cells through metabolic acidification, GAPDH-dependent translational control and NFAT inhibition, yet under certain conditions lactate oxidation in T cells can support memory formation and long-term antitumor responses [25]. Sun and colleagues compile evidence that lactylation can either promote or restrain tumor progression depending on the modified substrate and cellular context: in gastric cancer, lactylation of NBS1 and MRE11 enhances DNA repair and chemoresistance, whereas lactylation of METTL16 promotes cuproptosis. They also highlight that histone lactylation in CD8⁺ T cells can enhance effector functions and restrain tumor growth, underscoring the substrate- and cell type–specific nature of lactylation [177]. Taken together, these studies argue against a simple, uniformly pro-tumor view of lactate and lactylation. Their overall impact depends strongly on cell type (tumor versus immune), local metabolite and pH conditions, and the specific proteins and lysine residues that are modified.

Specific cancer systems and lactylation
Lactylation, although playing distinct roles across different cancer types, is increasingly recognized as a key player in controlling how tumors grow, evade the immune system, and spread to other tissues [178]. In the following sections, we will systematically explore the role of lactylation in the most prevalent cancers across different organ systems, aiming to identify novel therapeutic targets for future precision treatments (Table 2) (Fig. 6).

Lactylation and tumors of the digestive system

Esophageal cancer
Esophageal cancer (EC) typically arises from the cells lining the esophagus and is mainly classified into two subtypes: adenocarcinoma and squamous cell carcinoma. With an estimated 580,000 new cases annually worldwide, it ranks as the seventh most common cancer, while its approximately 510,000 deaths place it sixth in cancer-related mortality. These figures highlight the substantial threat EC poses to global health [206]. The main risk factors for EC involve genetic predisposition, lifestyle choices, dietary habits, alcohol use, tobacco smoking, and environmental exposures [207].
Recent studies link lactylation to poor prognosis in esophageal cancer (EC). In ESCC, lncRNA AP001885.4 is upregulated and predicts unfavorable outcomes; it promotes proliferation by increasing H3K18la and enhancing NF-κB (p65)-driven c-myc transcription, while also stabilizing c-myc mRNA via METTL3-dependent m6A. AP001885.4 further associates with LDHA/LDHB, suggesting a coupling between lactate metabolism and lactylation, and highlighting a metabolic–epigenetic–oncogene axis with therapeutic potential [179]. A recent study by Qiao et al. revealed that hypoxia-induced lactylation of serine hydroxymethyltransferase 2 (SHMT2) enhances its protein stability in EC cells, thereby promoting glycolysis, stemness, and tumor progression. Lactate accumulation under hypoxic conditions facilitated SHMT2 lactylation, which subsequently upregulated MTHFD1L expression, a key enzyme involved in mitochondrial one carbon metabolism. This upregulation contributed to increased proliferation, migration, and invasion of tumor cells. Silencing SHMT2 suppressed these malignant phenotypes, while reintroducing MTHFD1L restored them, suggesting that targeting the hypoxia-driven SHMT2 lactylation pathway may offer a promising therapeutic strategy in EC [180]. Hypoxia also elevates histone lactylation, with H3K9la enriched at the LAMC2 promoter to activate LAMC2 transcription, which in turn stimulates PI3K/Akt signaling and VEGFA expression to drive invasion, metastasis, and angiogenesis; inhibiting lactate production (e.g., oxamate) suppresses H3K9la and LAMC2, underscoring lactate-dependency of this axis [181].

Gastric cancer
Gastric cancer (GC) is a major global malignancy with high incidence and mortality [208]. In 2020, about 1.09 million new cases and 769,000 deaths were reported worldwide; although overall incidence is declining, increases have been noted in some regions and in adults younger than 45 years [209]. Established risk factors include Helicobacter pylori infection, smoking, alcohol intake, obesity, and high-salt diets [210]. Recent work emphasizes metabolic reprogramming in GC: metabolomics shows elevated glycolytic intermediates, including lactate and pyruvate, in tumors versus adjacent tissues, alongside activation of myo-inositol–linked lipid biosynthesis, consistent with enhanced glycolysis and lipogenesis and potential vulnerability to glycolytic inhibition [211, 212].
Recent studies have highlighted the significant involvement of lactylation in the progression of GC. In GC, copper elevates K229 lactylation of METTL16 (via AARS1/2), enhancing its m⁶A activity to stabilize FDX1 and promote cuproptosis (via DLAT lipoylation and proteotoxic stress). SIRT2 acts as a delactylase that removes K229la. Combining elesclomol with the SIRT2 inhibitor AGK2 synergistically triggers cuproptosis and suppresses tumor growth in vitro and in vivo. These findings highlight lactylation of non-histone proteins as a critical epigenetic switch in metabolic stress-induced cell death and propose METTL16-K229 lactylation as a promising therapeutic target in GC [182]. GLUT3 is upregulated in primary/metastatic GC and correlates with LDHA and lactylation pathways. GLUT3 increases glycolytic flux and lactate, elevating histone H3K9la/H3K18la/H3K56la, which drives epithelial–mesenchymal transition (EMT), proliferation, migration, and invasion. GLUT3 knockdown lowers LDHA and histone lactylation and attenuates EMT, whereas LDHA overexpression rescues these effects; xenografts confirm this regulatory axis. This study underscores the interplay between metabolic reprogramming and epigenetic modification, establishing GLUT3 as a promising diagnostic marker and therapeutic target in GC [96]. TME lactate induces H3K18la, transcriptionally upregulating VCAM1, which activates AKT/mTOR, promoting proliferation, EMT, and migration. VCAM1 also induces CXCL1 to recruit hGC-MSCs and M2 macrophages, fostering immunosuppression and metastasis; high VCAM1 associates with poor prognosis. These findings highlight the critical role of the H3K18la–VCAM1–AKT/mTOR–CXCL1 cascade in driving GC progression and immune evasion, providing new insights into potential therapeutic strategies [183].

Colorectal cancer
Colorectal cancer (CRC) remains the third most common malignancy and the second leading cause of cancer mortality worldwide [213]. CRC accounts for over 1.85 million newly diagnosed cases and around 900,000 deaths annually on a global scale [214]. CRC is a multifactorial disease, with modifiable risk factors such as high intake of processed or red meat, low consumption of dietary fiber, fruits, and vegetables, along with smoking, excessive alcohol use, and lack of physical activity [215].
Recent studies indicate that lactylation contributes to CRC initiation and progression. In CRC, HDAC1 lactylation at K412 (HDAC1K412la), driven by elevated lactate in the TME, reinforces ferroptosis resistance by repressing H3K27ac at the FTO and ALKBH5 promoters, maintaining high m6A on FSP1 mRNA and stabilizing the ferroptosis suppressor FSP1. HDAC inhibitors (e.g., SAHA, TSA) reduce HDAC1K412la, increase FTO/ALKBH5, promote m6A demethylation and degradation of FSP1, and thereby sensitize CRC cells to ferroptosis in vitro and in vivo [184]. In colorectal cancer stem cells, glycolysis-derived lactate induces p300-dependent H4K12la, while HDAC1 erases this mark; H4K12la activates GCLC transcription to boost glutathione, limit lipid peroxidation, suppress ferroptosis, and drive chemoresistance. Targeting LDHA, p300, or GCLC restores ferroptosis sensitivity and enhances oxaliplatin efficacy [42]. In CRC liver metastasis, GPR37 promotes glycolysis via Hippo–YAP1, upregulates LDHA, increases H3K18la, and elevates CXCL1/CXCL5 to recruit neutrophils and create an immunosuppressive microenvironment; genetic or pharmacologic inhibition (Hypocrellin B) reduces metastasis and synergizes with anti–PD-1 therapy in vivo [185].

Hepatocellular carcinoma
Hepatocellular carcinoma (HCC) accounts for approximately 75% of liver cancers and remains a major global health burden [216]. In 2022, liver cancer caused 866,136 new cases and 758,725 deaths worldwide, ranking sixth for incidence and third for cancer-related mortality [217]. HCC arises from chronic hepatic injury and inflammation driven by multiple etiologies, with HBV and HCV infection remaining the leading global risk factors and accounting for more than half of cases [218]. Other established contributors include alcohol use, aflatoxin B1 exposure (especially in synergy with HBV), and the growing impact of metabolic dysfunction–associated steatotic liver disease (MASLD/NAFLD) and NASH, with additional region-dependent roles for smoking, hereditary disorders (e.g., hemochromatosis), and environmental hepatotoxins/air pollutants [219, 220].
Lactylation is increasingly implicated in HCC initiation, progression, and therapy resistance. In oxaliplatin/5-FU–resistant HCC, enhanced glycolysis elevates lactate and H3K14la, which induces NEDD4 to ubiquitinate and degrade PTEN, thereby activating PI3K–AKT–mTOR signaling, further reinforcing glycolysis and sustaining multidrug resistance; inhibiting lactylation or restoring PTEN reverses resistance in vitro and in vivo [95]. Tumor–stroma interactions also shape lactylation programs: single-cell and spatial transcriptomics identify distinct HSC subsets with active lactylation and stromal–tumor crosstalk, and nominate AKR1B10 as a lactylation-associated prognostic factor that promotes H3K18la and HCC growth, whereas AKR1B10 silencing suppresses proliferation and perturbs survival pathways [186]. In addition, H3K18la enrichment at the HECTD2 promoter upregulates this E3 ligase to drive KEAP1 K48-linked ubiquitination, NRF2 activation, and reduced lenvatinib cytotoxicity; clinically, high HECTD2 correlates with worse prognosis and poor lenvatinib response, while nanoparticle delivery of siHECTD2 restores drug sensitivity in HCC models [187].

Pancreatic cancer
Pancreatic cancer (PC), known for its exceptionally high mortality rate, ranks as the seventh leading cause of cancer-related deaths globally [221]. About 90% of PC is described as pancreatic ductal adenocarcinoma (PDAC) [222]. PC risk rises with hereditary mutations, chronic pancreatitis, pancreatic cystic lesions, and diabetes [223]. Nearly 80% of patients are first diagnosed with locally advanced or metastatic disease, effectively removing the chance for curative resection. Overall survival at five years is roughly 10%, and even among early‑stage cases eligible for surgery, the 5‑year survival rate remains under 31% [224].
Growing evidence indicates that lactate drive PC progression by coupling metabolic rewiring to epigenetic regulation via lactylation. In PDAC, elevated lactate increases H3K18la, which activates TTK and BUB1B transcription; these mitotic regulators further enhance p300 expression and LDHA phosphorylation, reinforcing glycolysis and lactylation in a positive feedback loop. Disrupting glycolysis or silencing LDHA reduces H3K18la and suppresses tumor growth and metastasis in vitro and in vivo, supporting the glycolysis–H3K18la–TTK/BUB1B axis as a therapeutic target [90]. Another study identified ENSA-K63la as a glycolysis-driven lactylation event that triggers the SRC–STAT3–CCL2 pathway, promoting recruitment and immunosuppressive polarization of TAMs, limiting CD8⁺ T-cell infiltration, and contributing to immune checkpoint blockade resistance; targeting ENSA-K63la/CCL2 (or KRAS) reprograms the immune microenvironment and improves immunotherapy responses [188]. Under glucose deprivation, lactate also sustains ductal adenocarcinoma (PAAD) cell survival by inducing EP300-dependent NMNAT1-K128la, which enhances NMNAT1 nuclear localization/activity to maintain nuclear NAD⁺ salvage and viability via Sirt1; pan-lysine lactylation is elevated in PAAD and associates with poor prognosis, and this axis promotes tumor growth in KPC and xenograft models [189].
Across many digestive system tumors, lactylation can be framed as a metabolic–epigenetic adaptation switch that translates lactate-rich states into coordinated programs of invasion/metastasis, immune escape and treatment tolerance. A recurring pattern is transcriptional activation associated with histone lactylation, complemented by non-histone lactylation that modulates RNA metabolism and protein stability/ubiquitin signaling—supporting the view of lactylation as a modular integrator rather than a single marker. However, much of the evidence remains correlative or relies on upstream metabolic perturbations, and lactylation-specific causality can be confounded by shared acylation machinery, co-existing lysine marks and assay specificity limits. Key gaps include in vivo site-causal validation, cell-type/spatial attribution of lactate sources and lactylation programs, and quantitative stoichiometry/therapeutic windows needed for actionable biomarkers and rational combinations.

Lactylation and tumors of the respiratory system

Lung cancer
Lung cancer represents a significant global public health challenge, ranking as the foremost cause of cancer-related deaths and the second most commonly diagnosed malignancy. Non–small cell lung cancer (NSCLC) accounts for nearly 85% of all lung cancer cases [225, 226]. Lung cancer is a multifactorial disease with both environmental and genetic risk factors contributing to its pathogenesis. Tobacco smoking remains the most significant etiological factor, accounting for over 80% of all lung cancer cases globally; the risk increases with both the duration and intensity of exposure [227]. However, a growing proportion of lung cancers, particularly adenocarcinomas, now arise in never-smokers, especially among women and individuals of East Asian descent [228]. In these populations, environmental exposures such as indoor radon gas, secondhand smoke, and ambient air pollution, especially fine particulate matter (PM2.5), have been identified as important contributing factors [229]. Genetic predispositions, such as germline mutations in EGFR and polymorphisms in loci like 5p15 (TERT), have also been implicated, especially in familial or early-onset cases [230].
Lactylation is increasingly recognized as a key epigenetic mechanism in lung cancer that links glycolytic lactate to immune evasion, tumor aggressiveness, and metastatic programs. In NSCLC, glycolysis-derived lactate drives H3K18la enrichment at the POM121 promoter, enhancing MYC nuclear translocation and PD-L1 (CD274) transcription to suppress cytotoxic T lymphocyte (CTL) activity; inhibiting glycolysis or knocking down LDHA/LDHB reduces H3K18la and PD-L1, restores CD8⁺ T-cell function, and synergizes with anti–PD-1 therapy in vivo [190]. Chang et al. demonstrated that NCAPD3 is significantly upregulated in lung cancer and associated with poor overall survival. Functional assays revealed that NCAPD3 promotes cell proliferation, migration, invasion, and inhibits apoptosis in lung cancer cells. Mechanistically, NCAPD3 enhances aerobic glycolysis by activating the MEK/ERK signaling pathway and upregulating LDHA expression. Moreover, lactate-induced histone lactylation, particularly H3K18la, was found to elevate NCAPD3 levels, suggesting a positive feedback loop between glycolysis, lactylation, and NCAPD3 expression. In vivo, NCAPD3 knockdown suppressed tumor growth and reduced lactylation and MEK/ERK activation. These findings identify NCAPD3 as a lactylation-responsive oncogene that links metabolic reprogramming to epigenetic regulation and tumor progression in lung cancer [191]. Lung cancer stem cells further remodel metastatic potential via exosomal Mir100hg, which enhances glycolysis (via ALDOA) to raise lactate and induce H3K14la at promoters of metastasis-associated genes (e.g., LMAN1, OSTC, P4HA1), activating pro-metastatic transcriptional programs and increasing colonization in vivo [192]. Yan et al. discovered that exposing NSCLC cells to 1% oxygen intensifies glycolytic flux, raises intracellular lactate, and markedly increases global Kla; under these conditions SOX9 becomes the key lactylated transcription factor, its modification stabilizing the protein and activating gene programmes that confer stem-like properties, migration, and invasion. Disabling glycolysis with 2-deoxy-D-glucose or silencing SOX9 eliminates these malignant traits in vitro and sharply limits xenograft growth in vivo, underscoring that the hypoxia-induced glycolysis–lactate–SOX9 lactylation cascade sustains tumor aggressiveness and positions both lactate production and SOX9la as promising therapeutic targets in NSCLC [193].
In lung cancer, lactylation is better viewed as a chromatin imprint written by the combined forces of hypoxic niches and immune selection pressure, rather than as a readout of a single pathway. Its significance lies in linking the spatial gradients typical of pulmonary tumors (oxygen, nutrients and acidity) to a persistent set of cellular states shaped by therapy, particularly immunotherapy: on one side, tumor cells in lactate-rich, low-oxygen regions acquire heightened adaptability and plasticity; on the other, the TME is remodeled toward a T cell–unfavorable “cold” state. In parallel, exosome-driven metabolic rewiring suggests that lactylation-associated traits can propagate across tumor subpopulations, offering a plausible explanation for metastatic plasticity and intratumoral heterogeneity in lung cancer. Importantly, much of the current literature still relies on inhibiting glycolysis or lowering lactate as the main causal evidence, interventions that simultaneously perturb multiple metabolic and epigenetic axes and may directly impair immune-cell metabolism, making it difficult to attribute phenotypes specifically to lactylation. Key translational gaps therefore include determining which lactylation events are truly required drivers in vivo rather than by-products, resolving the spatial and directional relationships among lactate sources, lactylation deposition and immunosuppressive/invasive phenotypes, and establishing quantitative thresholds and stratification criteria to enable rational combination strategies (for example with immunotherapy or anti-metastatic regimens) without broadly compromising anti-tumor immune metabolism.

Lactylation and tumors of the hematopoietic system

Leukemia
Leukemia comprises a heterogeneous group of hematologic malignancies characterized by uncontrolled proliferation of abnormal leukocytes, commonly classified by disease tempo (acute versus chronic) and lineage (myeloid versus lymphoid) [231]. Acute myeloid leukemia (AML) is an aggressive stem/progenitor cell malignancy in which rapid blast expansion suppresses normal hematopoiesis, causing cytopenia-related complications; diagnosis is established by the presence of blasts in blood or bone marrow [232, 233]. Acute lymphoblastic leukemia (ALL) arises from lymphoid progenitors and is relatively uncommon, with an incidence of 1.1 cases per 100,000 individuals annually in the United States [234]. Chronic lymphocytic leukemia (CLL), the most common adult leukemia, is defined by at least 5 × 10⁹/L monoclonal B cells with a characteristic immunophenotype (CD5, CD19, dim CD20, CD23); many patients are initially asymptomatic, and treatment is typically deferred until iwCLL 2018 criteria are met, with risk stratification guided by the CLL-IPI (including TP53 disruption, IGHV status, β₂-microglobulin, stage, and age) [235]. Chronic myeloid leukemia (CML) is driven by the Philadelphia chromosome BCR::ABL1 fusion, whose constitutively active tyrosine kinase underpins leukemic proliferation and progression [236]. Established risk factors include chemical exposures (such as benzene), prior chemo- or radiotherapy, inherited predisposition syndromes, antecedent hematologic disorders, family history, and demographic factors [237].
Lactylation is increasingly recognized as an important epigenetic and post-translational regulator in leukemia, shaping transcriptional programs, metabolic adaptation, and immune evasion. In AML, STAT5 promotes immune escape through a lactate-driven mechanism: STAT5 activation enhances glycolysis and elevates intracellular lactate, which facilitates nuclear translocation of E3BP and increases histone H4K5 lactylation at the PD-L1 promoter, thereby upregulating PD-L1 expression. Clinically, STAT5 levels correlate with lactate concentration and PD-L1 expression in bone marrow samples, and genetic disruption of PD-L1 or PD-1 blockade can restore CD8⁺ T-cell function, supporting this STAT5–E3BP–lactylation axis as a potential biomarker and therapeutic target for improving responses to immune checkpoint inhibition in AML [167]. In KMT2A-rearranged AML, integrative multi-omics analyses identified a six-gene lactylation-associated signature (PFN1, S100A6, CBR1, LDHB, LGALS1, and PRDX1) that distinguishes this high-risk subtype from cytogenetically normal AML and associates with immune infiltration and metabolic pathway activity, defining molecular subtypes with distinct immunometabolism dependencies and suggesting potential sensitivity to PI3K, MEK, and proteasome inhibitors in signature-high cases [194]. In pediatric ALL, dysregulated sphingomyelin metabolism, characterized by increased SGMS1 expression and reduced SMPD3 expression, activates a HIF-1α/GLUT1 glycolytic program that elevates lactate and induces lysine 14 lactylation of procaspase-3, preventing its cleavage and thereby disabling apoptosis; interventions that rebalance sphingomyelin metabolism reduce lactate, remove this modification, restore caspase-3 activation, and suppress leukemia growth in xenograft model [195].

Lymphomas
Lymphomas are heterogeneous malignancies arising from clonal expansion of B-, T-, or NK-cell lineages across differentiation stages [238]. The 5th edition WHO Classification of Haematolymphoid Tumors recognizes more than 100 clinicopathologic entities, broadly comprising Hodgkin lymphoma and diverse non-Hodgkin lymphomas defined by integrated morphology, immunophenotype, and genomic features [239]. Globally, non-Hodgkin lymphoma accounted for approximately 545,000 new cases and 260,000 deaths in 2020, and modeling studies estimate a lifetime risk of 1.67% [240, 241]. Lymphomagenesis reflects convergent genetic, infectious, immune, environmental, and host factors. Recurrent genetic lesions include hallmark translocations such as t(14;18) leading to BCL2 overexpression in follicular lymphoma and MYC rearrangements in Burkitt lymphoma, as well as dysregulated signaling via NF-κB, JAK/STAT, and PI3K/AKT pathways [242]. Oncogenic infections, particularly Epstein–Barr virus, and also hepatitis C virus and human T-lymphotropic virus-1, contribute to lymphoma development in selected entities [243–245]. Risk is further increased in settings of chronic immune stimulation or immunosuppression, including autoimmune disease and transplant-associated immunosuppression [246], and epidemiologic studies implicate chemical exposures such as pesticides and benzene in elevated non-Hodgkin lymphoma risk [247].
Protein lactylation has been increasingly linked to lymphoma biology through effects on transcriptional control, immune contexture, and tumor progression. In T-cell lymphoma, Yu et al. analyzed lymph node biopsies from 70 de novo patients and 22 reactive hyperplasia controls and found markedly elevated histone lactylation, particularly H3K9la, which correlated with higher LDH levels and poorer prognosis. Glycolysis inhibition (2-DG) or LDH inhibition (oxamate) reduced lactate and H3K9la, induced G0/G1 arrest, and suppressed TCL proliferation. Integrated RNA-seq and ChIP-seq identified SFXN1 as a key H3K9la-activated target, with promoter enrichment of H3K9la and functional evidence that SFXN1 supports TCL growth in vitro and in vivo, nominating an LDH–H3K9la–SFXN1 metabolic–epigenetic axis as a therapeutic vulnerability [248]. In diffuse large B-cell lymphoma, Zhu et al. integrated TCGA and GEO transcriptomes with machine learning to derive a seven-gene lactylation-related signature (CHERP, DHX9, EMG1, HNRNPH1, LCP1, RPS11, UBE2E1) that stratified patients by survival risk; the RiskScore associated with adverse outcome, macrophage infiltration, and predicted drug sensitivity differences. Mechanistically, HNRNPH1 knockdown increased lactate and global lactylation, enhanced proliferation, and reduced apoptosis in vitro and in vivo, and patient immunohistochemistry linked higher lactylation with increased Bcl-2, C-myc, and P53 expression and metabolic activity (LDH, SUVmax). Together, these data support dysregulated lactylation as a prognostic and potentially actionable axis in DLBCL [249].

Multiple myeloma
Multiple myeloma (MM) is an incurable, genetically heterogeneous plasma-cell malignancy that ranks as the second most common haematological cancer, presenting with clonal bone-marrow infiltration and monoclonal immunoglobulin production [250]. MM represents an increasingly significant global health challenge, with an estimated 188 000 new cases and 121 000 deaths reported in 2022; age-standardised incidence rates are highest in North America and Australasia (≥ 4 per 100 000 population), and demographic modelling predicts that, in the absence of effective preventive strategies, worldwide MM incidence and mortality will rise by 71% and 79%, respectively, by 2045 [251]. From an aetiological standpoint, nearly all MM cases evolve from the precursor entities monoclonal gammopathy of undetermined significance (MGUS) and smouldering MM; whole-genome sequencing of more than 1 000 patient samples has mapped a sequential accrual of hallmark lesions—including canonical immunoglobulin heavy-chain translocations (e.g., t(4;14)/MMSET, t(11;14)/CCND1), hyperdiploidy, chromosome-1q amplifications and activating KRAS/NRAS mutations—that together constitute an ‘MM-like’ genomic signature predictive of malignant progression [252]. Large-scale epidemiological and meta-analytic studies show that MM risk rises with advancing age, male sex, African ancestry, first-degree familial history of plasma-cell disorders, elevated body-mass index, and prolonged occupational exposure to ionising radiation, benzene, pesticides or other organic solvents [253, 254].
Recent work links Kla to MM pathogenesis and drug resistance. Guo et al. generated a lenalidomide-resistant MM model and, using quantitative proteomics, identified extensive glycolysis-associated increases in lactylation, including 1,241 lysine-lactylation sites. Pharmacologic inhibition of glycolysis (2-deoxy-D-glucose or oxamate) reduced global lactylation and restored lenalidomide sensitivity, supporting a lactate-fueled resistance mechanism. Resistant cells showed coordinated activation of survival pathways and reduced expression of several CRL4–CRBN components, indicating both CRBN-dependent and -independent escape routes. Importantly, chromatin analyses revealed enrichment of H4K8la at promoters such as CDK6 and ECHS1, promoting their transcription independently of acetylation and implicating a glycolysis–H4K8la axis as a therapeutic vulnerability in lenalidomide-resistant MM [255]. Sun et al. profiled lactylation-related genes (LRGs) in MM and developed a nine-gene prognostic signature (TRIM28, PPIA, SOD1, RRP1B, IARS2, RB1, PFN1, PRCC, FABP5) that stratified patients into risk groups with distinct overall survival across multiple cohorts; a nomogram incorporating the LRG score with clinical variables improved prognostic performance. Pathway analyses suggested divergent metabolic and immune features between risk groups, and functional validation highlighted PFN1 as a pro-myeloma factor whose knockdown suppresses proliferation and induces apoptosis, supporting LRG-based stratification for personalized management [256].
Hematopoietic malignancies offer a particular setting to understand lactylation, because leukemia and lymphoma cells grow within bone marrow or lymphoid tissues where tumor cells, stromal cells and immune cells are in close contact and share the same metabolic environment. In this context, lactate-linked lactylation may help malignant cells coordinate growth demands with immune escape and stress tolerance. Recent studies suggest several recurring themes. In AML, lactate-associated histone lactylation has been linked to PD-L1 transcription, consistent with an immunosuppressive state that may be relevant to checkpoint-based strategies. In ALL, lactylation can act closer to cell fate control by modifying apoptotic machinery and limiting caspase activation, providing a direct route to apoptosis resistance. In lymphomas, histone lactylation has been associated with activation of metabolic support pathways, including one-carbon metabolism, whereas in multiple myeloma it is connected to epigenetic upregulation of pro-survival programs and acquired drug resistance that is not fully explained by the canonical CRBN pathway alone. Multi-omics signatures related to lactylation further point to “lactylation-high” subgroups with distinct immunometabolic features and potential treatment sensitivities. At the same time, much of the evidence is still based on upstream metabolic perturbations or association analyses, and it remains unclear which lactylation events are required in vivo and which reflect broader metabolic stress. Key next steps are to validate the most informative sites and substrates in patient-relevant models, map lactylation across malignant, stromal and immune compartments in clinical samples, and define quantitative thresholds that can support reproducible stratification and rational combination treatment.

Lactylation and tumors of the nervous system

Glioblastoma
Gliomas comprise a major fraction of primary brain tumors, with glioblastoma representing the most common malignant subtype in adults and remaining highly lethal despite multimodal therapy [257, 258]. Contemporary diagnosis integrates histology with molecular biomarkers as defined by the 2021 WHO classification of CNS tumors, which organizes gliomas into major clinicopathologic entities across adult- and pediatric-type diffuse gliomas, circumscribed astrocytic gliomas, and related ependymal tumors [259]. Established risk factors are limited: increasing age is associated with higher incidence, and ionizing radiation remains the only confirmed environmental exposure [260]; inherited cancer predisposition syndromes, including NF1 [261], tuberous sclerosis complex [262], and Li–Fraumeni syndrome [263], also confer elevated glioma risk.
Glioma development occurs within a lactate-rich microenvironment, and lactylation provides a plausible molecular bridge linking this metabolic landscape to immune escape and adaptive survival. In glioblastoma, lactate derived from glioma stem cells and tumor-associated microglia enhances histone lactylation to upregulate the “don’t-eat-me” signal CD47 and suppress microglial phagocytosis; CBX3 acts as a reader that cooperates with EP300 to amplify lactylation and immunosuppressive signaling, and inhibiting lactate production (e.g., dichloroacetate) synergizes with anti-CD47 therapy in vivo [124]. Hypoxia-driven extracellular lactate can also be imported into macrophages via MCT1, increasing macrophage histone lactylation (notably H3K18la) and activating TNFSF9 transcription to reinforce an M2-like, tumor-promoting phenotype; glycolysis inhibition or MCT1 silencing reverses this immunosuppressive polarization and limits glioma growth [196]. In the setting of chemoresistance, chronic temozolomide exposure elevates lactate and H3K9la, which is deposited by EP300 at the LUC7L2 promoter; increased LUC7L2 drives aberrant MLH1 splicing and loss of mismatch repair, whereas genetic disruption of LUC7L2 or lowering lactate (LDHA/LDHB knockdown or the brain-penetrant LDH inhibitor stiripentol) reduces H3K9la and resensitizes tumors to temozolomide [197]. Myeloid programs further converge on lactylation-dependent immunosuppression: tumor signals activate PERK–ATF4 to induce GLUT1-high glycolysis in macrophages, generating lactate that promotes H3K18la at the IL10 promoter and robust IL-10 production; blocking PERK, glycolysis/LDH, or p300 reduces lactylation, restores CD8⁺ T-cell infiltration, and improves responses when combined with T-cell agonism [198]. Finally, an NF-κB–driven circuit links the Warburg effect to H3K18la at the LINC01127 promoter, inducing this lncRNA to enhance MAP4K4–JNK signaling and reinforce stem-like traits and immunosuppressive features; disrupting lactate production, lactylation, or JNK signaling suppresses intracranial tumor growth and prolongs survival [199].
In glioma, lactylation stands out as a mechanism that links the tumor’s glycolytic lifestyle to the unique CNS immune landscape, with particularly strong effects on microglia and infiltrating macrophages. Rather than operating as a uniform “mark,” lactylation in glioblastoma tends to appear in discrete, pathway-defining nodes that reshape how myeloid cells sense and respond to tumor-derived metabolites, and how tumor cells maintain stem-like states and survive therapeutic stress. This creates an ecology in which metabolic cues are converted into stable immune suppression and adaptive resistance programs, helping explain why lactate-rich regions often coincide with poor immune clearance and recurrence. Future work should treat lactylation in glioma as a brain-specific, dynamic program rather than a static marker. Priorities include defining the in situ biochemical flux from lactate to lactylating capacity using isotope tracing, mapping when lactylation is reversible during therapy, and testing whether CNS lactate shuttling linked to neural activity shapes tumor and myeloid lactylation states. It is also important to distinguish lactylation dependencies between resident microglia and recruited macrophages, and to integrate lactylation with glioma genotype and chromatin background, including IDH status, to identify truly “lactylation-addicted” tumors. Translation will require brain-feasible biomarkers and interventions, ideally pairing lactylation signatures with repeatable clinical readouts such as metabolic imaging and ensuring adequate CNS drug delivery.

Lactylation and tumors of the endocrine system

Thyroid cancer
Thyroid cancer is the most common endocrine malignancy [264]. Based on estimates from GLOBOCAN 2022, thyroid cancer accounted for approximately 821,214 new diagnoses and 47,507 deaths worldwide, underscoring the malignancy’s accelerating global incidence [265]. The 2022 fifth edition of the WHO Classification of Endocrine and Neuroendocrine Tumors refined thyroid neoplasia into benign, low-risk and malignant follicular-cell lineages—introducing “high-grade differentiated carcinoma” and salivary-gland–like variants to align histology with genomic insights [266]. Ionizing radiation remains the clearest etiologic driver: a recent systematic review of occupational low-dose exposure (< 10 mSv) still demonstrated a dose-responsive excess risk of thyroid cancer [267]. Large-scale population studies have shown that being female, having obesity, and certain modifiable lifestyle factors—such as low physical activity and exposure to second-hand smoke—are associated with a higher risk of developing thyroid cancer, suggesting that some cases may be preventable through lifestyle changes [268],
Research on lactylation in thyroid cancer has been continuously advancing, offering new insights into its potential roles in tumor progression and therapeutic targeting. In an anaplastic thyroid cancer (ATC) model, Wang et al. showed that oncogenic BRAFV600E rewires glycolysis to accumulate lactate, which the acetyl-transferase p300 installs as H4K12 lactylation (H4K12la) at gene promoters, thereby activating a proliferative transcriptional program. Pharmacologic BRAF inhibition with vemurafenib decreased lactate production, reduced H4K12la occupancy at cell-cycle loci such as CDK1 and CCNE1, and blocked tumor cell growth, whereas exogenous lactate or p300 activation restored the mark and partially rescued proliferation. Importantly, restricting lactylation—via glycolytic blockade, p300 inhibition, or lactate scavenging—synergized with vemurafenib to produce sustained tumor regression in 8505c xenografts with no added systemic toxicity, highlighting the glycolysis–lactylation axis as a druggable vulnerability that can sensitize BRAFV600E-positive ATC to targeted therapy [200].

Neuroblastoma
Neuroblastoma is an embryonal malignancy arising from sympathoadrenal neural-crest precursors and is the most common extracranial solid tumor of childhood, accounting for roughly 8–10% of all paediatric cancers yet about 15% of childhood cancer-related deaths [269]. Neuroblastoma is thought to arise when sympathoadrenal neural-crest precursors fail to complete normal specification and, under early driver hits such as MYCN amplification, activating ALK mutations or telomerase-enhancer rearrangements, acquire malignant potential [270]. Although the majority of neuroblastoma cases are sporadic, approximately 1–2% occur in familial clusters, typically associated with highly penetrant germline mutations in ALK (activating mutations) or PHOX2B (loss-of-function variants), which are recognized as the principal predisposition genes for this malignancy [271, 272].
The role of lactylation in neuroblastoma is gradually being elucidated through ongoing research. Sun et al. revealed that down-regulation of the HSP40 co-chaperone DNAJC12 reprograms neuroblastoma metabolism toward heightened glycolysis, thereby elevating intracellular lactate, which in turn amplifies p300-mediated histone H4K5la and diminishes SIRT2-dependent delactylation. Chromatin immunoprecipitation demonstrated a marked enrichment of H4K5la at the COL1A1 promoter, driving its transcription and promoting extracellular-matrix remodeling, actin reorganization, and aggressive tumor proliferation and invasion. Restoring DNAJC12 expression or pharmacologically inhibiting glycolysis, p300 activity, SIRT2 modulation, or COL1A1 expression mitigated these malignant phenotypes. Clinically, low DNAJC12 or high H4K5la levels in patient tumors correlated with significantly inferior overall survival, identifying the DNAJC12–lactate–H4K5la–COL1A1 circuit as both a prognostic marker and a promising therapeutic vulnerability in high-risk neuroblastoma [273]. Zu et al. showed that the cytosolic and nuclear deacetylase SIRT2 is a bona fide histone delactylase, efficiently removing Kla from multiple histone substrates and markedly lowering H3K18la, H4K8la, and H4K12la levels in neuroblastoma cells. Loss of SIRT2, or expression of a catalytically inert H187Y mutant, increases global and site-specific lactylation—most prominently at H4K8la—which in turn activates oncogenic transcripts such as SERPING1 and TRPV4, thereby accelerating proliferation and migration in SH-SY5Y cells; re-expression of wild-type SIRT2 or pharmacological inhibition of lactate production reverses these effects. Mass spectrometric profiling of 88 primary tumors further reveals that low SIRT2 or high SERPING1 expression is associated with significantly poorer overall survival, underscoring the tumor-suppressive role of the SIRT2-H4K8la axis in high-risk neuroblastoma. Collectively, these findings identify SIRT2-mediated delactylation as a critical metabolic and epigenetic checkpoint and suggest that activating SIRT2 or curbing lactate-driven H4K8la may offer therapeutic benefit in neuroblastoma [71].

Lactylation and tumors of the urogenital system

Renal cell carcinoma
Renal cell carcinoma (RCC) encompasses a hetero geneous group of cancers derived from renal tubular epithelial cells [274]. RCC accounts for roughly 90% of malignant kidney tumors, with clear-cell histology comprising about three quarters of cases and papillary and chromophobe subtypes making up most of the remainder [275]. Loss of VHL in clear-cell RCC and activating alterations in MET in papillary RCC illustrate the distinct molecular drivers that underpin these principal subtypes [276, 277]. Smoking shows a positive dose–response association with RCC risk: even light smokers face a significant increase in incidence, whereas sustained abstinence markedly attenuates that risk [278]. Obesity and hypertension promote renal tumorigenesis through mechanisms involving insulin-like growth factor signaling, chronic inflammation, and altered renal hemodynamics, and are consistently identified across multiple cohort studies as among the most important modifiable risk factors for RCC [279].
Histone and non-histone lactylation link glycolytic lactate to epigenetic/epitranscriptomic reprogramming in RCC, promoting pro-tumor transcriptional programs, immune evasion, and therapy resistance. Wan et al. generated the first lysine-lactylome of ccRCC, reporting a tumor-specific increase in global Kla and mapping 5,725 Kla sites across 1,620 proteins; differential analysis identified 441 altered sites on 239 proteins. Kla was enriched on cytoplasmic/mitochondrial proteins and clustered in central metabolic pathways (glycolysis/TCA, nucleotide metabolism), with lactylated hub enzymes (e.g., GAPDH, LDHA, ATP5F1A/B) anchoring a dense metabolic interaction network, positioning Kla as an organized regulator of carbon metabolism and a potential therapeutic vulnerability in ccRCC [280]. Mechanistically, Dai et al. showed that hypoxia-driven lactate accumulation elevates Kla and highlighted YTHDC1 K82 as a hyper-lactylated site installed by p300 and erased by HDAC3. K82 lactylation enhances YTHDC1 liquid–liquid phase separation, stabilizing oncogenic transcripts (including BCL2 and E2F2) by protecting them from decay, thereby promoting proliferation, invasion, and sunitinib resistance; the non-lactylatable K82R mutant suppresses these phenotypes, and xenograft data support a causal role in tumor growth and metastasis. These results define a lactate–p300–YTHDC1 axis and nominate p300 inhibition, LLPS disruption, or delactylase activation as RCC-directed strategies [201]. In parallel, Yang et al. reported that VHL loss increases intracellular lactate to drive H3K18la-dependent activation of the PDGFRB promoter, forming a feed-forward loop in which PDGFRβ signaling reinforces glycolysis and histone lactylation. Inhibition of LDHA/lactylation (oxamate) or PDGFRβ (axitinib) suppressed malignant phenotypes and slowed PDX growth, with the combination producing the strongest tumor regression and reduced pulmonary metastasis, supporting co-targeting of lactylation and PDGFRβ in VHL-mutant ccRCC [97].

Breast cancer
Breast cancer is the most commonly diagnosed malignancy worldwide and a leading cause of female cancer mortality [281]. Globally, 2022 saw roughly 2.3 million newly diagnosed cases and about 670,000 associated deaths [282]. In the United States alone, 310,720 invasive and 56 500 in-situ cases are anticipated in 2024, with 42,250 fatalities, giving women a lifetime risk of about one in eight [283]. Molecular profiling classifies breast cancer into four intrinsic subtypes: luminal A, luminal B, HER2-enriched, and basal-like or triple-negative; each subtype possesses distinct biological characteristics, prognostic implications, and therapeutic strategies [284]. Major breast-cancer hazards include advancing age and high-penetrance germline BRCA1/BRCA2 mutations, which elevate lifetime risk roughly five- to eight-fold [285].
Recent evidence links lactylation to breast cancer progression and clinically relevant heterogeneity. Li et al. showed that lactate accumulation in triple-negative breast cancer (TNBC) promotes histone H4K12 lactylation; in 60 paired specimens, higher H4K12la associated with advanced stage and shorter relapse-free survival. CUT&Tag mapped H4K12la to the SLFN5 promoter, where it repressed this tumor-suppressive program and reduced apoptosis, while inhibiting lactate production or preventing K12 modification restored SLFN5 expression and restrained xenograft growth, supporting a lactate–H4K12la–SLFN5 axis as a potential vulnerability in TNBC [286]. Jiao et al. profiled 22 lactylation-related genes across 1,420 breast cancers, identifying two immunometabolic clusters with distinct outcomes; they further derived a seven-gene signature (RAD51, CASP14, NEK10, PCP2, IDO1, CLSTN2, IGHG1) that stratified patients by risk and correlated with immune features and predicted drug sensitivities, suggesting lactylation signatures may inform prognosis and treatment selection [287]. Cui et al. performed lactyl-enriched quantitative proteomics in paired TNBC tissues and identified 58 Kla sites on 48 proteins, with H4K12lac showing the strongest tumor-specific increase; tissue microarrays confirmed prevalent nuclear H4K12lac staining and associations with plasma lactate, Ki-67, and inferior survival, with high expression remaining an independent adverse prognostic factor. Together, these studies nominate H4K12 lactylation as a metabolically linked epigenetic marker and functional driver in TNBC [288].

Ovarian cancer
Ovarian cancer remains the most lethal gynaecologic malignancy [289]. Genomic and pathological data indicate that most high-grade serous tumors, the predominant ovarian-cancer histology, arise from serous tubal intra-epithelial carcinoma in the distal fallopian tube rather than from the ovarian surface epithelium [290]. Inherited pathogenic variants in BRCA1 and BRCA2 remain the strongest drivers, conferring lifetime risks of roughly 44 percent and 17 percent, respectively [285]. A positive family history or Lynch-syndrome mutations provides an additional two- to three-fold increase [291]. Late detection, drug resistance, heterogeneity, and scarce salvage options keep ovarian-cancer outcomes poor [292].
In ovarian cancer, lactate-driven lactylation supports both immunometabolism remodeling and therapy resistance. One study linked higher circulating lactate to advanced FIGO stage and poorer prognosis, and identified a metabolically active subtype in which macrophage GPR132 signaling promotes H3K18la enrichment at the CCL18 promoter, driving an M2-like phenotype and high CCL18 secretion that enhances tumor proliferation, migration, and liver metastasis; GPR132 knockdown or CCL18 neutralization reversed these effects [293]. In platinum-resistant disease, lactate accumulation is accompanied by GCN5-dependent H3K9la and RAD51 K73 lactylation, which upregulate RAD51/BRCA2, strengthen homologous recombination repair, and reduce cisplatin sensitivity; genetic or pharmacologic GCN5 inhibition (MB-3) removes these marks and resensitizes cell-line and patient-derived xenografts to platinum [294]. A separate study showed that Tanshinone I suppresses FOXK1/2-driven glycolysis, lowers lactate, and reduces H3K18la-associated oncogenic transcription; exogenous lactate restores H3K18la, and Tan-I treatment slows xenograft growth while shifting the microenvironment toward higher CD8⁺ T cells and M1 macrophages with fewer M2 macrophages and Tregs [295].

Cervical cancer
Cervical cancer remains a major global health problem, accounting for an estimated 662,301 incident cases and 348,874 fatalities in 2022, the burden falling disproportionately on low- and middle-income countries [296]. Virtually all tumors arise from persistent infection by oncogenic human papillomavirus, chiefly HPV-16 and HPV-18, which together drive around 60% of squamous-cell and adenocarcinoma cases [297]. Cervical cancer treatment is often hindered by late-stage diagnosis and the emergence of therapeutic resistance [297].
In cervical cancer, glycolysis-derived lactate drives tumor progression through both tumor-intrinsic and microenvironmental lactylation programs. One study identified a lactate–DCBLD1–G6PD axis in which lactate increases DCBLD1 transcription via HIF-1α and induces DCBLD1 K172 lactylation, stabilizing DCBLD1 by preventing ubiquitin-mediated degradation; stabilized DCBLD1 protects G6PD from autophagic turnover, boosts oxidative PPP flux, elevates NADPH/glutathione, reduces ROS, and promotes proliferation and motility. Genetic or pharmacologic inhibition of DCBLD1 or G6PD suppresses tumor growth in vitro and in xenografts [202]. In an HPV-driven mechanism, HPV16 E6 promotes cervical-cancer metabolism by removing G6PD K45 lactylation, enhancing G6PD dimerization and activity to increase PPP flux and redox buffering; restoring a lactylation-mimetic state or inhibiting G6PD (6-aminonicotinamide) restrains tumor growth in vivo [203]. In the TME, lactate also programs macrophage immunosuppression: lactate uptake via MCT1 increases H3K18la at the GPD2 promoter, elevating GPD2 expression and enforcing M2 polarization, while lactate-blocking strategies or GPD2 deletion reverse this phenotype [298].
Urogenital tumors show some lactylation patterns that are fairly distinctive. In renal cancer, lactylation often plugs directly into the major driver biology of the disease (such as VHL loss and hypoxia signaling) and can form self-reinforcing loops with growth-factor pathways and targeted-drug response. In several urogenital cancers, lactylation also seems to act beyond chromatin, affecting RNA handling and nuclear organization (for example through m6A-related factors and phase-separated condensates). Another recurring theme is its connection to redox control and biosynthesis, particularly via the pentose phosphate pathway and NADPH/glutathione balance, which is especially clear in HPV-related cervical cancer. Immune effects are frequently macrophage-centered, where lactate sensing helps maintain M2-like programs, and in breast cancer, site-specific histone lactylation has been reported as a usable prognostic readout in aggressive subtypes. Going forward, urogenital tumors should be studied in ways that match their defining biology and treatment pressures. In RCC, the priority is to map how lactylation is embedded in the VHL–hypoxia network and growth-factor signaling, and whether it helps explain adaptation to VEGFR/MET-directed therapy. In ovarian cancer, the key question is how lactylation strengthens homologous-recombination repair and thereby shapes platinum and PARP-inhibitor response. In cervical cancer, lactylation needs to be examined alongside HPV-driven metabolic rewiring, especially its link to PPP-dependent redox buffering. In breast cancer, particularly TNBC, it remains important to clarify when H4K12 lactylation is a true driver of transcriptional silencing versus a correlate of high lactate. Finally, because these tumors repeatedly connect lactylation to RNA handling and nuclear condensates, targeting lactylation-dependent RNA regulation may offer a more selective entry point than broad glycolysis blockade.

Lactylation and tumors of the integumentary system

Cutaneous melanoma
Cutaneous melanoma is an aggressive malignancy of melanocytes with a substantial global burden. GLOBOCAN 2020 estimated about 325,000 new cases and 57,000 deaths worldwide, with the highest incidence in Australia/New Zealand and Western Europe [299]. In the United States, the American Cancer Society projected approximately 105,000 new invasive cases and more than 8,400 deaths in 2025 [227]. Genomically, BRAF V600 mutations occur in about half of tumors, while NRAS and NF1 alterations define additional clinically relevant subsets [300]. Despite therapeutic advances, unmet needs include primary or acquired resistance to PD-1 blockade [301], limited durability of combined BRAF/MEK inhibition due to MAPK pathway reactivation [302], poor outcomes after brain metastasis [303], and lower checkpoint responses in mucosal and acral melanoma than in cutaneous [304].
Lactylation has been implicated in melanoma progression and treatment resistance through both histone and non-histone mechanisms. In BRAFV600E melanoma, Li et al. showed that relapse-associated glycolytic rebound increases lactate and induces LSD1 K503 lactylation, which stabilizes LSD1 and reshapes its chromatin targeting. Lactylated LSD1 cooperates with FosL1 to repress the transferrin receptor, limiting iron uptake and protecting drug-tolerant cells from ferroptosis; pharmacologic LSD1 inhibition restores ferroptotic vulnerability and, when combined with anti–PD-1 therapy, produces strong antitumor effects in resistant xenograft models [204]. In ocular melanoma, Yu et al. reported elevated global Kla and H3K18la in patient tumors, with higher H3K18la associated with worse outcomes. Reducing lactate production lowers H3K18la and suppresses tumor growth, while mechanistic work identified EP300-dependent H3K18la at the YTHDF2 promoter, increasing YTHDF2 and promoting decay of tumor-suppressive PER1 and TP53 transcripts, thereby facilitating malignancy [205]. In cutaneous melanoma, integrative analyses further identified CALML5 as a lactylation-linked prognostic factor, with higher tumor expression associated with poorer survival and an immune-cold microenvironment, supporting lactylation-related networks as potential biomarkers and immunomodulatory targets [305].
In melanoma, lactylation is increasingly viewed as an adaptive response that becomes more prominent under therapeutic pressure and in lactate-rich niches. Relapse after MAPK-targeted therapy is often accompanied by glycolytic rebound and lactate accumulation, conditions that can promote lactylation-driven rewiring of chromatin regulation and survival programs in drug-tolerant cells. A distinctive aspect is that both histone and non-histone lactylation have been implicated, linking lactate to transcriptional control as well as to protein stability and stress tolerance, including altered iron handling and ferroptosis susceptibility. Overall, lactylation tends to track with aggressive behavior, recurrence risk, and an immune-suppressed tumor state, providing a plausible bridge between targeted-therapy resistance and poor immunotherapy responses in melanoma.

Lactylation and tumors of the musculoskeletal system

Osteosarcoma
Osteosarcoma remains the most common primary malignant bone tumor in adolescence [306]. Therapy for osteosarcoma has reached a plateau. While limb-sparing surgery combined with methotrexate, doxorubicin, and cisplatin achieves five-year survival rates of 60 to 70 percent in patients with localised disease, those who present with or relapse to pulmonary metastases experience markedly poorer outcomes, with survival dropping to 20 to 30 percent. This stagnation is driven by the tumor’s strong propensity for early lung dissemination, the emergence of both intrinsic and acquired resistance to chemotherapy, and a highly immunosuppressive tumor micro-environment that significantly limits the efficacy of ICIs and other immunotherapeutic approaches [307].
Studying lactylation in osteosarcoma may uncover vulnerabilities linked to chemoresistance and immune regulation. Peng et al. analyzed TARGET-OS and GSE21257, identified 36 prognosis-associated lactylation-related genes, and stratified 141 cases into two molecular groups. Tumors with low lactylation signatures showed higher immune–stromal scores and markedly better survival, whereas lactylation-high tumors were immunologically “cold” and associated with early death. A 13-gene lactylation score further separated outcomes, with low-score cases showing substantially improved overall survival, while high-score tumors captured most metastatic events and exhibited low expression of PD-1, PD-L1, and CTLA-4. Drug-sensitivity modeling suggested distinct therapeutic liabilities across groups, nominating different targeted classes for lactylation-low versus lactylation-high disease [308]. Wang et al. integrated single-cell RNA-seq (GSE152048) with bulk cohorts (TARGET, GSE16088) and identified a lactate-dependent malignant subset characterized by coordinated activation of glycolytic and fatty-acid programs. This state localized to an immune-infiltrated but functionally exhausted niche with checkpoint enrichment, including PD-L1, shaped by cytokine signaling networks. Multi-model feature selection highlighted NDUFAF6 as a lactate-regulated hub whose expression increased along malignant pseudotime, correlated with a more immune-suppressed phenotype, and predicted in-silico sensitivity to several candidate agents. Functionally, NDUFAF6 knockdown reduced 143B cell viability and invasiveness under lactate exposure, and immunohistochemistry confirmed elevated NDUFAF6 in patient tumors. Together, these data link lactate/lactylation-associated metabolic rewiring to immune escape in osteosarcoma and support metabolism-guided therapeutic strategies [309].
In osteosarcoma, lactylation is most often captured as a transcriptome-level state that stratifies patients by outcome. Lactylation-high tumors tend to associate with poorer survival and higher metastatic risk, and they frequently show an immune-cold microenvironment with lower immune–stromal scores and weaker checkpoint-related signals. Single-cell data further suggest marked metabolic heterogeneity, including lactate-dependent malignant subsets coupled to distinct immune niches. Overall, lactylation-linked programs in osteosarcoma appear to track with both metastatic potential and immune context, supporting their use for risk stratification and metabolism-guided combination strategies.

Cross-system commonalities and differences
Across diverse malignancies, lactylation converges on a set of recurring themes. First, H3K18la is frequently associated with transcriptional activation of pro-tumor programs and adverse phenotypes, as initially shown in the discovery study of histone lactylation and subsequently reported in multiple cancer contexts [83, 178, 310]. Second, tumors with high MCT4 (SLC16A3) expression exhibit elevated lactate export and are often linked to immunosuppressive microenvironments and reduced immunotherapy efficacy; conversely, blocking MCT4 can improve leukocyte infiltration and T-cell activation, supporting the functional link between lactate handling and antitumor immunity [311–313]. From a mechanistic perspective, lactate remodels the TME metabolically and epigenetically by acting as an oxidative fuel and by supplying the substrate for histone lactylation across tumor, stromal, and immune compartments [314]. At the same time, system-specific differences are evident: lactylation targets and downstream pathways vary by lineage and context (e.g., hypoxia, pH, MCT1/4 balance), and immune effects can be bidirectional. For example, lactate can under certain conditions enhance CD8⁺ T-cell stemness and antitumor function. Beyond histones, emerging evidence for non-histone lactylation adds another cross-system layer by modulating protein function in signaling, immune regulation, and genome maintenance, potentially amplifying context dependence across tissues [315]. These nuances underscore that while the lactate–lactylation axis provides a unifying framework, its phenotypic outputs depend on tissue ecology and co-occurring PTMs [175].

Clinical trials in cancer therapy
Among current intervention points, it is important to distinguish between targeting lactylation itself and targeting the broader lactate biology that feeds into lactylation. Accordingly, p300/CBP, LDH and MCTs are best viewed as clinically druggable entry points into the lactate–lactylation axis rather than lactylation-specific drug targets: p300/CBP are pleiotropic co-activators that write multiple acyl-lysine marks, LDH inhibition alters redox balance and central carbon flux, and MCT blockade reshapes extracellular acidity and metabolite exchange across tumor and immune compartments [23, 48, 124, 316, 317]. Therefore, phenotypes observed with these interventions cannot be attributed to lactylation without dedicated evidence that changes in relevant Kla marks track with the proposed transcriptional and functional outputs.Within this framework, p300/CBP remain attractive nodes because, as writers of Kla and acetylation, they can couple glycolytic flux to chromatin and reinforce pro-tumor gene programmes; pharmacologic inhibition may dampen lactate-associated chromatin states in selected settings [318]. Complementing this epigenetic lever, controlling lactate production and transport is often mechanistically informative: LDH inhibition can reduce lactate generation while constraining NAD⁺ recycling and glycolytic throughput [316], whereas MCT inhibition can limit transmembrane lactate exchange, reduce extracellular acidosis and potentially attenuate lactate-conditioned immunosuppressive phenotypes [317]. In parallel, inhibiting p300/CBP halts the chromatin writing of lactate signals, dampening pro-survival and immune-evasive transcriptional programs. Co-targeting LDH/MCTs and p300/CBP therefore collapses both the metabolic source and the nuclear signal, yielding direct antitumor effects and sensitizing tumors to immunotherapy, chemotherapy, and thermotherapy by improving effector T-cell function, drug penetration, and stress-response control. Conceptually, co-targeting lactate handling (LDH/MCTs) together with chromatin writing (p300/CBP) may be rational in lactate-high tumors or therapy-induced “lactate rebound” contexts, but it should be framed as a strategy to re-shape lactate-driven transcriptional and microenvironmental states, supported by orthogonal pharmacodynamic readouts (for example, site-specific Kla quantified by mass spectrometry or rigorously validated antibodies).
Beyond conceptual proposals to block lactate production or erase lactylation, clinically relevant strategies already encompass repurposed metabolic drugs, direct inhibitors of lactate-handling enzymes and transporters, and epigenetic modulators of the lactylation machinery. Biguanides such as metformin, which inhibit mitochondrial complex I and shift the NAD +/NADH balance, indirectly lower glycolytic flux and lactate accumulation and are being evaluated in multiple phase II and III trials in combination with chemo-, radio- and immunotherapies [319, 320]. In addition, several small molecules targeting lactate-generating or lactate-exporting proteins, including LDH inhibitors [321] and selective MCT1 blockers exemplified by AZD3965 [39], have advanced into early-phase clinical testing in lymphoma and solid tumors, where they aim to collapse lactate-fueled metabolic circuits and remodel the acidic TME. These developments underscore that pharmacological targeting of lactate and lactylation is already entering the clinic and provide a framework for positioning lactylation-focused therapies within broader metabolic oncology (Table 3).

Targeting p300/CBP
Although lactylation has only recently emerged as a novel epigenetic modification, mounting preclinical evidence has illuminated its critical role in regulating tumor metabolism, immune evasion, and therapy resistance. Lactylation, primarily catalyzed by enzymes such as p300/CBP and derived from intracellular lactate, modulates gene expression programs linked to glycolysis, angiogenesis, and immunosuppression [322]. These findings have sparked growing interest in pharmacologically targeting p300/CBP to disrupt lactylation-associated tumor phenotypes. In this context, the clinical-grade small-molecule inhibitor CCS1477 (Inobrodib) represents a compelling therapeutic candidate. CCS1477 selectively targets the bromodomains of p300 and CBP, disrupting their chromatin association and downstream transcriptional programs. Although it was originally developed to suppress oncogenic transcription factors such as androgen receptor and MYC, recent findings have shown that CCS1477 also effectively represses NRF2-dependent cytoprotective gene expression, which plays a central role in chemoresistance and immune evasion in KEAP1-mutant cancers [323]. Given that NRF2-driven transcription relies heavily on p300/CBP recruitment to enhancer regions, CCS1477 treatment not only blocks acetylation-dependent transcriptional activation but may also indirectly attenuate histone lactylation, thereby disrupting lactate-fueled epigenetic signaling. Currently, CCS1477 is being evaluated in two Phase I/IIa clinical trials (NCT03568656 and NCT04068597), involving patients with advanced solid tumors and hematologic malignancies, including NSCLC characterized by constitutive NRF2 activation. These trials aim to assess the safety, pharmacokinetics, and anti-tumor efficacy of CCS1477 both as monotherapy and in combination with standard chemotherapeutics or immunomodulators. As p300/CBP inhibition may concurrently suppress both acetylation and lactylation marks, CCS1477 provides a mechanistically innovative approach to epigenetic therapy in cancer, with potential to remodel tumor metabolic-epigenetic circuits and overcome drug resistance [1, 324].
Beyond CCS1477, several additional CBP/p300-directed agents have now reached clinical or near-clinical development and may serve as first-in-class pharmacologic tools to modulate Kla. Dual BET–CBP/p300 inhibitors such as NEO2734 (EP31670), which has shown potent activity against acute myeloid leukaemia and prostate cancer models, exemplify this class [325]. Catalytic HAT inhibitors exemplified by A-485 provide highly selective chemical probes that reduce histone H3 and H4 acylation and have demonstrated efficacy in several tumor types in vivo, including androgen receptor–driven prostate cancer and pituitary tumors [326, 327]. Although these compounds were originally developed to block histone acetylation, their ability to dampen CBP/p300-dependent acylation marks at promoters and enhancers raises the possibility that catalytic CBP/p300 inhibitors might also act as indirect anti-lactylation agents, especially in settings where p300/CBP contribute substantially to histone lactylation. Complementing writer inhibition, the concept of pharmacologically activating delactylases such as SIRT3 is emerging. Although selective SIRT3 activators are currently confined to preclinical studies, accumulating evidence indicates that SIRT3 can de-lactylate both mitochondrial and nuclear substrates, highlighting a potential route to therapeutically erase pathological lactylation [328].

Targeting LDHs
LDH, a tetrameric enzyme, catalyzes the reversible conversion of pyruvate and lactate. In cancer cells, the predominant isoform LDHA enables evasion of OXPHOS, diverting pyruvate-derived intermediates into the pentose phosphate pathway to sustain proliferation [329, 330]. Elevated LDHA expression is associated with unfavorable outcomes in multiple cancers [331, 332]. LDHB overexpression occurs in various cancers, including breast, thyroid, lung, and pancreatic, and is strongly linked to poor prognosis [333, 334]. Current research aims to develop LDH inhibitors with improved cellular potency, pharmacokinetics, and selectivity, though most remain at the preclinical stage. AT-101 (gossypol), originally explored as an EGFR mutation–targeted agent, has entered phase I/II randomized trials in advanced non-small cell lung cancer, head and neck cancer, and metastatic castration-resistant prostate cancer. In combination with standard chemotherapy, AT-101 has shown potential benefits in high-risk patients and in subsets with prolonged progression-free or overall survival (NCT01003769, NCT00286780, NCT00988169, NCT00540722) [335]. Nonetheless, AT-101, its derivative FX-11, galloflavin, and N-hydroxyindole–based compounds are potent cell-active LDHA inhibitors, but their therapeutic use has yet to reach clinical application [336, 337]. Intravenous NCI-006, an LDHA/B inhibitor, and oral GNE-140, an LDHA-selective inhibitor, both suppress LDH activity and tumor growth in PC mouse models [338, 339]. Moreover, the LDH PROTAC degrader MS6105 induces time- and ubiquitin–proteasome–dependent degradation of LDHA/B, thereby inhibiting proliferation in multiple PC cell line [340]. The efficacy of LDHA inhibitors is hindered by intertumoral LDH heterogeneity and isoform switching driven by metabolic reprogramming [341].
LDH-targeted therapy can also be effective in combination with adjuvant treatments, and growing evidence supports its role in enhancing the efficacy of immunotherapy [342]. In a humanized NSCLC mouse model, combining oxamate with the PD-1 inhibitor pembrolizumab enhanced CD8⁺ T-cell infiltration, suppressed tumor growth, and increased immunotherapy sensitivity [343]. In a melanoma mouse model, the LDHA inhibitor GSK2837808A improved the therapeutic effectiveness of adoptive T-cell therapy (ACT) [344]. Similarly, inhibition of LDH has been shown to potentiate photothermal therapy (PTT). Zhao et al. developed a zinc-enriched nanosystem incorporating the glycolysis inhibitor LND and the LDHA inhibitor zinc to achieve combined glycolysis modulation and PTT. Additionally, oxamate-induced LDHA inhibition promoted ROS accumulation and ATP depletion, causing DNA damage, impairing DNA repair, and improving radiotherapy efficacy in NSCLC [345]. Moreover, LDHA inhibition can restore radioiodine (RAI) sensitivity in PTC. Shi et al. revealed that the lncRNA glycine-rich long non-coding transcript (GLTC) impedes SIRT5 binding to LDHA, preventing K155 succinylation. This suppression of LDHA activity reduces tumor progression and RAI resistance in PTC [346]. The results indicate that the presence of free zinc ions inhibits LDHA activity in a concentration-dependent manner and increases LDH efflux, thereby enhancing the efficacy of PTT and synergistically suppressing primary melanoma and lung metastases [347].

Targeting MCTs
Targeting MCTs markedly disrupts metabolic symbiosis [348]. A wide spectrum of MCT inhibitors has been developed, ranging from classical agents such as CHC [349, 350], organomercurial compounds, and photothialdehyde benzenesulfonate [351], to more selective second-generation inhibitors including AR-C155858 for MCT1/2 [352] and BAY8002 or SR13800 for MCT1 [353]. In addition, AstraZeneca’s MCT1/2 inhibitor AZD3965 has demonstrated promising preclinical efficacy in small cell lung cancer (SCLC) [354]. AZD3965 has also shown therapeutic efficacy in models of MCT1-positive Burkitt’s lymphoma [355], as well as breast and gastric cancers [356, 357]. AZD3965 has undergone a Phase I/II clinical trial (NCT01791595) in solid tumors and diffuse large B-cell lymphoma, which assessed its pharmacokinetics and safety profile, demonstrating tolerability at doses sufficient for target engagement [39, 355]. The dose-limiting toxicities were on-target, largely dose-dependent, and manifested as asymptomatic, reversible ocular changes. Preclinical data and retrospective analyses indicate that MCT4 can compensate for MCT1 loss when MCT1 is downregulated. These findings highlight the complexity of MCT targeting and the potential for resistance mechanisms [354]. Moreover, MCT1 inhibition enhances CD8⁺ T-cell tumor reactivity by modulating lactate catabolism. Beyond MCT1 inhibition, several older metabolic agents provide additional proof-of-principle that lactate metabolism can be pharmacologically manipulated in patients. Dichloroacetate, a pyruvate dehydrogenase kinase inhibitor that promotes oxidative metabolism, has been evaluated in early-phase trials in advanced solid tumors (e.g. NCT01029925), where it reduced blood lactate levels and altered tumor glucose uptake patterns. Likewise, the glycolytic inhibitor 2-deoxy-D-glucose has been tested in combination with chemoradiotherapy (NCT00096707), demonstrating acceptable safety and evidence of on-target metabolic effects. Although these studies did not measure protein lactylation, their ability to modulate systemic and intratumoral lactate underscores the feasibility of targeting lactate supply as an upstream handle on the lactylation machinery. MCT4 inhibitors also show therapeutic promise, with MCT4 expression in hypoxic TMEs driven by HIF-1α [358]. Under hypoxia, MCT4 knockdown re-sensitizes lung adenocarcinoma cell lines to inhibitors of glycolysis and OXPHOS, highlighting its importance for lactate-targeted therapy [359]. MCT4i is a promising therapeutic choice for GBM [360], lung adenocarcinoma [361], gastric cancer [362]. Combining the MCT1 inhibitor AZD3965 with the MCT4 inhibitor AZ93 markedly suppressed proliferation in CRC cell lines [363]. 7-Aminocarboxycoumarin (7ACC) compounds block both MCT1 and MCT4, preventing MCT1 compensation after MCT4 inhibition, downregulating mitochondrial pyruvate transport, causing intracellular pyruvate accumulation, and inhibiting compensatory lactate uptake. Expression of MCT1 and MCT4 at the plasma membrane is maintained through the chaperone proteins CD147 (basigin, BSG) [364]. In preclinical prostate cancer models, CD147/BSG inhibition modulates MCT1/MCT4-mediated lactate transport, reducing lactate efflux and suppressing tumor growth [365]. The CD147 dimerization blocker AC-73 [366], the human/mouse chimeric IgG1 antibody metuzumab [367],and the organomercurial p-chloromercuribenzene sulfonate (pCMBS) [368], which blocks MCT1/MCT4-CD147 binding 118 are available CD147-targeted anticancer drugs. Nonetheless, on a cautionary note, CD147 is ubiquitously expressed and interacts with other proteins at the cell surface. Thus, strategies selectively targeting CD147-MCT interactions ought to minimize drug toxicity and establish a therapeutic window [369]. Evidence suggests that combining MCT-targeted therapy with other treatments can enhance therapeutic outcomes. In a breast cancer mouse model, Li et al. showed that the MCT inhibitor syrosingopine reduced Treg cell numbers while increasing NK cells and M1-type TAMs, indicating a reversal of the immunosuppressive TME [370]. Ma et al. reported that lithium carbonate (LC) promotes MCT1 localization to mitochondrial membranes and facilitates lactate influx into mitochondria in vitro and in vivo. This enhanced energy supply reactivated tumor-reactive CD8⁺ T cells, thereby sensitizing immunotherapy against CRC, melanoma, and breast cancer [371]. This highlights the potential of MCT-targeted therapy to act synergistically with immunotherapy.

Other lactate-targeted strategies
Beyond inhibiting lactate production (LDH) and transmembrane transport (MCTs), and intercepting its epigenetic readout (p300/CBP), the lactate axis encompasses multiple additional druggable nodes across different regulatory layers. HCAR1 (GPR81) is a Gi-coupled receptor for lactate that is upregulated across multiple solid tumors (e.g., breast, cervical, pancreatic), where it promotes malignant phenotypes, including PI3K/AKT-dependent growth and angiogenesis [372]. Functionally, silencing GPR81 in breast cancer cells reduces lactate release, downregulates MCT4 (and in some contexts MCT1), diminishes glycolytic ATP production, and suppresses proliferation and anchorage-independent growth in vitro as well as tumor growth in vivo. Genetic knockout in MCF7 cells further shows a metabolic shift from glycolysis toward oxidative phosphorylation (ECAR↓/OCR↑) accompanied by reduced expression and activity of rate-limiting glycolytic enzymes (PFK, HK), underscoring HCAR1’s role in maintaining the glycolytic program. Beyond tumor-intrinsic effects, lactate engages GPR81 on dendritic cells to suppress MHC-II expression, highlighting a paracrine route to immune evasion; lactate–HCAR1 signaling also drives PD-L1 induction in tumor cells, reinforcing immunosuppressive circuits [373, 374]. Collectively, these data position HCAR1/GPR81 as a druggable lactate sensor that coordinates glycolytic metabolism, lactate handling, and immune suppression in TME. In addition, researchers have proposed lactate oxidase (LOx) as a therapeutic strategy to lower lactate levels, generate H₂O₂, and recruit immune cells, thereby overcoming immunosuppression and enhancing immunotherapy responsiveness [375, 376]. Moreover, LOx delivery systems have progressed from polymer nanocarriers to self-assembled nanoparticles, with these refinements expanding its potential for enhancing chemotherapy and sonodynamic therapy (SDT) [377, 378]. LOx-mediated oxidation of lactate generates pyruvate, thereby activating CRISPR/Cas9-mediated editing of signal-regulatory protein alpha (SIRPα). In combination with a metal–organic framework (MOF), LOx and these plasmids are assembled into LPZ nanoparticles (LOx, Cas9/sgSIRPα plasmids, mannose-modified PEG–loaded ZIF-67), promoting the polarization of M2 to M1 macrophages and suppressing tumor growth in in situ breast cancer models [379]. This approach enables LOx-triggered, CRISPR/Cas9-based editing of macrophages in the TME and may improve immunotherapy [380]. For example, Luo et al. employed nano-ZIF-8 as a carrier to develop the Hb-LOx-DOX-ZIF8@platelet membrane nanosystem (HLDZ@PM NPs), which effectively increased tumor sensitivity to DOX-based chemotherapy [381]. Anchoring LOx to the surface of lactobacillus (LA) enhanced lesion targeting and delivery efficiency, enabling complete lactate oxidation and intratumoral oxygen depletion, thereby activating chemotherapeutic agents to induce apoptosis [382]. Zhang et al. designed a metal–phenolic network nanocomplex integrating LOx with the mitochondrial respiration inhibitor atovaquone (ATO) to remodel the immunosuppressive TME, achieving greater therapeutic efficacy than either agent alone in breast cancer SDT [383].

Lactylation-related biomarkers
Since lactylation was proposed in 2019 as a histone modification linking metabolism and transcription, it has evolved into a biomarker repertoire with assay tractability and biological interpretability. At the tissue level, H3K18la in epithelial ovarian cancer (IHC, n = 112) associates with worse OS/PFS and is an independent risk factor; in NSCLC, H3K18la promotes immune evasion via the POM121–MYC–PD-L1 axis, suggesting predictive value for immunotherapy combinations [190, 384]. Beyond these entities, high nuclear histone lactylation has also been linked to larger tumor size, more advanced stage and significantly worse survival in ocular melanoma(IHC, n = 58) [205], and similar associations with aggressive molecular features and shorter progression-free survival have been reported in renal cell carcinoma (IHC, n = 65) [97], indicating that lactylation-rich tumors tend to follow a more malignant clinical course. At the circulating level, pancreatic cancer serum H3K18la correlates with lactate and CA19-9 and achieves a diagnostic AUC of approximately 0.85, though standardized case–control design is still needed, underscoring the importance of prospective validation [83]. At the transcriptomic level, lactylation-related gene (LRG) signatures built across multiple tumors stratify prognosis and, in some studies, associate with responses to targeted or immune therapies (e.g., HCC, gastric, pancreatic), complementing tissue and serum markers [385, 386]. Because lactylation depends on lactate flux, MCT4 (SLC16A3) can serve as an indirect “lactate-ecosystem” marker: pan-cancer analyses link high expression to poorer survival and immunosuppression, supporting its use for stratification in metabolism–immunity combinations [387]. In addition, multi-omics work in HCC nominates AKR1B10 as co-occurring with high lactylation activity and as prognostic with functional support, extending the “marker-plus-mechanism” evidence chain [186]. Taken together, H3K18la (tissue and serum), lactylation-related gene signatures, MCT4 and AKR1B10 outline a multi-level biomarker system for the lactate–lactylation axis with clear translational promise; however, current evidence is largely retrospective and exploratory, and genuine clinical readiness will require harmonized antibodies and scoring systems, predefined cut-offs and prospective validation in large, multi-center patient cohorts [80].

Future challenges and opportunities

Challenges and solutions
Although interest in lactylation has surged, several foundational gaps still limit mechanistic consolidation and clinical translation. The enzymatic machinery is incompletely defined. While p300/HBO1 and AARS1 have been identified as potential “writers,” and HDAC3 and SIRT2 as “erasers,” the full repertoire of lactylation writers/erasers/readers has yet to be elucidated, particularly given the overlap of these enzymes with other short-chain acylations [388]. In parallel, the relative contributions and biological significance of L- versus D-derived lactyl marks in tumors remain insufficiently resolved [47]. Measurement is another constraint: while lactylation can be detected by mass spectrometry and tissue assays, broadly accepted, stereochemically aware workflows and clinically validated biomarkers are still lacking; for example, H3K18la has been associated with poor outcome in epithelial ovarian cancer, but requires deeper validation across cohorts and settings [388, 389]. Moreover, dependence on lactate metabolism and lactylation varies across tumor types, molecular subtypes, and stages, as illustrated by subtype-specific metabolic programs in breast cancer, and the temporal ordering of lactylation events between tumor and immune compartments remains unclear [178, 390]. Finally, lactylation operates within a dense PTM and metabolic network, complicating attribution of phenotypes to lactylation per se and leaving open whether it is oncogenic or tumor-suppressive in specific contexts [31, 178, 388].
Despite these constraints, several practical translational anchors are already available. Validated writers and erasers can be quantified using in-vitro reconstitution with site-resolved proteomics and extended to chromatin-relevant contexts [43, 44, 59, 71, 78], and patient-oriented readouts can combine quantitative LC–MS with standardized tissue assays and, where feasible, noninvasive measures of lactate flux [52, 88, 384, 391, 392]. Lactate transport inhibition has also entered early-phase clinical testing, providing an actionable path for biomarker-enabled studies [15, 39]. Building on this evidence base, future work should remain deliberately focused on standardizing stereochemically controlled measurement pipelines, establishing causal writer/eraser/reader–substrate relationships and their effector consequences with orthogonal perturbations, and mapping heterogeneity through cross-cohort Kla–subtype atlases that connect MCT1/MCT4 balance, LDH isoforms, and lactylation fingerprints to clinically actionable subgroups and companion diagnostics [78, 88, 393, 394]. Early clinical studies can then adopt metabolically defined enrichment strategies and a small set of pre-specified pharmacodynamic endpoints linking target engagement to intratumoral lactate and tissue lactylation, while evaluating mechanism-based combinations, including immunotherapy, within adaptive designs [39].

Future opportunities
Despite these challenges, lactylation research holds substantial therapeutic promise in oncology. First, novel target discovery and drug development. The identification of new lactylation regulators has opened avenues for drug design. Examples include AARS1-mediated p53 lactylation [54], MOESIN lactylation modulating TGF-β signaling [165], and Mettl3 lactylation driving m6A-dependent immune suppression [166]. Targeting these “writers” or “readers” with small molecules could allow direct modulation of lactylation. Preclinical studies show that LDHA or MCT inhibition can reduce lactate accumulation and enhance immunotherapy efficacy; LDHA inhibitors combined with anti-PD-1 antibodies, for example, have demonstrated synergistic antitumor effects in HCC and glioblastoma [395, 396]. Small molecules targeting lactylation marks—such as dimethyl malonate (DML), royal jelly acid (RJA), and luteolin—have been shown to suppress H3K9la/H3K14la in liver and lung cancers [396–398]. Second, combination therapy and immune modulation. Lactylation plays a pivotal role in tumor immune evasion, and its inhibition can potentiate immunotherapy. For example, LDHA inhibition combined with CAR-T or anti-PD-1 therapy enhances tumor killing and reduces immunosuppressive molecule expression [396]. Strategies integrating metabolic inhibitors, epigenetic modulators, and immunotherapies may mitigate resistance and improve therapeutic outcomes. Third, biomarker and detection technology advancement. Emerging high-throughput approaches—including next-generation sequencing, single-cell transcriptomics, and ChIP-seq—enable systematic identification of lactylation regulators and target genes. Development of specific antibodies and mass spectrometry platforms could facilitate accurate lactylation quantification in patients, supporting treatment monitoring and prognostic stratification [399, 400]. Fourth, precision medicine and targeted delivery. Tumor heterogeneity necessitates individualized lactylation-targeting strategies. Nanocarriers or biomaterials enabling tumor-specific delivery of lactylation inhibitors can enhance bioavailability while minimizing systemic toxicity [401, 402] (Fig. 7). Integrating metabolic and epigenetic profiles into predictive models could pave the way for personalized therapeutic regimens [403].
Lactylation, as a novel link between metabolism and epigenetics, is rapidly emerging as a key player in cancer biology and therapy. The major current challenges include incomplete understanding of its enzymatic machinery, lack of robust detection methods and biomarkers, tumor-type heterogeneity, cross-talk with other modifications, and barriers to clinical translation. However, opportunities lie in the discovery of novel targets, expansion of combination treatment strategies, advances in detection technologies, and integration into precision medicine. By dissecting the regulatory networks of lactylation, developing specific modulators, and validating their clinical utility, lactylation may become a transformative avenue in cancer therapy.

Conclusion
In this comprehensive review, we systematically summarized the molecular mechanisms, functional roles, and clinical implications of protein lactylation in cancer. We first introduced lactylation as a novel post-translational modification that links cellular metabolism—especially aerobic glycolysis—to gene expression regulation. We detailed the enzymatic and non-enzymatic pathways that mediate Kla, and explored its impacts across various tumor types. These include promoting aerobic glycolysis (Warburg effect), modulating immune evasion through macrophage and T cell reprogramming, enhancing glutaminolysis and lipid biosynthesis, and conferring resistance to ferroptosis and chemotherapy. Furthermore, we presented recent evidence showing how lactylation remodels TME and interacts with other epigenetic marks, such as acetylation, methylation, and ubiquitination, to regulate cancer progression. By dissecting lactylation’s role across digestive, respiratory, hematologic, and nervous system tumors, our review provides a panoramic and mechanistic understanding of this epigenetic modification.
The significance of this work lies in its integrative perspective. Unlike prior reviews that focused on isolated cancers or limited molecular mechanisms, we offer the first pan-cancer synthesis of lactylation biology, emphasizing its dual role in tumor-intrinsic reprogramming and tumor–microenvironment interaction. Although lactate and lactylation were initially viewed mainly as pro-tumoral and immunosuppressive, accumulating evidence now supports a dual, context-dependent model in which they can either promote or restrain tumor growth and T-cell-mediated immunity, depending on the cellular compartment, metabolic niche and specific substrates modified. Recognizing this double-edged nature will be essential for designing therapeutic strategies that selectively exploit the beneficial aspects of lactate and lactylation while minimizing their tumor-promoting effects. We also identify druggable targets within lactylation pathways—including LDHA, p300/CBP, AARS1/2, and monocarboxylate transporters—and highlight their translational potential in guiding combination therapies, biomarker development, and nanomedicine strategies. As such, this review not only bridges cancer metabolism with epigenetics, but also advances lactylation as a unifying conceptual and therapeutic axis in oncology.
Looking forward, further investigations are needed to clarify the specificity, reversibility, and dynamics of lactylation across different tumor contexts and immune states. Despite significant progress in understanding the biological roles of lactylation, many fundamental questions remain unanswered. Future studies should aim to uncover the reversible regulatory mechanisms governing lactylation. While lactate-driven histone lactylation has been widely reported, the identification and functional characterization of its specific “erasers” (delactylases) are still in their infancy. Elucidating how these enzymes are regulated and how they modulate cellular phenotypes could provide critical insight into the dynamic nature of this post-translational modification.In addition, the causal relationship between lactylation and tumor drug resistance warrants further investigation. Although correlations have been observed in several cancers, direct mechanistic evidence is needed to determine whether lactylation actively contributes to chemoresistance or simply reflects metabolic rewiring. Functional studies employing CRISPR screening, lactylation mimetics/inhibitors, and resistant tumor models are essential to dissect this potential causality.Another critical direction is exploring the crosstalk between lactylation and other PTMs such as acetylation, methylation, and ubiquitination. This interplay may shape the chromatin landscape, influence gene expression programs, and modulate immune evasion or therapeutic responses in a highly context-specific manner. To translate these findings into clinical practice, future work should also focus on the development of specific inhibitors or activators of lactylation-related enzymes, and the assessment of lactylation-based biomarkers in large, well-characterized clinical cohorts. The integration of multi-omics strategies—including transcriptomics, proteomics, metabolomics, and epigenomics—will be crucial to map lactylation-mediated regulatory networks and identify actionable molecular signatures. Ultimately, the convergence of metabolic, epigenetic, and immunological research around lactylation is poised to open new avenues for precision oncology, offering novel strategies to overcome drug resistance, enhance the efficacy of immunotherapies, and reprogram the tumor ecosystem for durable clinical benefit.