Lipid metabolic reprogramming in colorectal cancer: Insights to mechanisms and therapeutics

Abstract

Colorectal cancer (CRC) exhibits profound lipid metabolic reprogramming, a hallmark of malignant transformation that supports tumorigenesis, immune evasion, and therapeutic resistance. Dysregulated lipid metabolism in CRC involves altered fatty acid synthesis, uptake, oxidation, and cholesterol metabolism, which collectively drive cancer cell proliferation, metastasis, and interactions with the tumor microenvironment (TME). This review synthesizes current insights into lipid metabolic rewiring in CRC, its role in shaping immunosuppressive TME dynamics, and emerging therapeutic strategies targeting lipid pathways.

Key Words: Colorectal cancer; Lipid metabolism reprogramming; High-fat diet; Tumor microenvironment; Targeted drugs

Core Tip: Lipid metabolic reprogramming is a hallmark of colorectal cancer, influencing tumor initiation, immune evasion, and therapeutic resistance. Targeting key nodes in lipid synthesis, oxidation, and transport, particularly in combination with immunotherapy, offers a transformative approach to colorectal cancer treatment. While preclinical models highlight the efficacy of metabolic-immune combinatory regimens, future studies should prioritize clinical translation of these strategies while addressing metabolic heterogeneity and microenvironmental crosstalk.

INTRODUCTION

Cancer cells exhibit profound metabolic reprogramming that supports their rapid proliferation, survival, and adaptation to diverse microenvironments. Among the most extensively studied metabolic alterations is the Warburg effect[1]. Warburg initially reported that cancer cells preferentially utilize glycolysis for adenosine triphosphate (ATP) generation, even under aerobic conditions. Research has further elucidated the critical roles of dysregulated metabolism involving glucose, amino acids, and lipids in cancer pathogenesis[2].

In addition to increased glucose uptake and aerobic glycolysis, lipid metabolism has emerged as a critical pathway in cancer biology, particularly in colorectal cancer (CRC) pathogenesis, in recent years[3]. Compared with normal cells, cancer cells exhibit significant alterations in lipid metabolism. A groundbreaking metabolomics study revealed significant metabolic reprogramming, including lipid reprogramming, in CRC samples and reported concurrent alterations across multiple biochemical categories[4]. These changes include increased de novo lipogenesis, increased fatty acid (FA) uptake, and altered FA oxidation (FAO)[5]. Lipid metabolism dysregulation alters membrane composition, gene expression, and signaling pathway activity, which directly impacts downstream cellular functions and promotes cancer initiation and progression[6]. Specific lipogenic enzymes [e.g., acetyl-CoA carboxylase 1 (ACC1) and FA synthase (FASN)] and related metabolic regulators are frequently dysregulated in multiple malignancies, including CRC[7].

Furthermore, despite their role in mediating the biological characteristics of tumor cells, substantial evidence suggests that lipids influence the function and status of immune cells within the tumor microenvironment (TME)[8]. The cells of the TME include tumor-associated macrophages (TAMs), tumor-infiltrating lymphocytes (TILs) T cells, dendritic cells (DCs), myeloid-derived suppressor cells (MDSCs), natural killer (NK) cells and cancer-associated fibroblasts (CAFs)[9]. Specifically, lipid metabolic reprogramming in these cells within the TME, which manifests as enhanced lipid uptake, accumulation, or FAO, drives the TME toward an immunosuppressive state, thereby facilitating tumor progression[10]. For example, regulatory T cells (Tregs) and TAMs exhibit increased lipid uptake and de novo synthesis, which facilitates their ability to promote CRC progression[11,12]. Similarly, cluster of differentiation (CD) 36 promotes the uptake of oxidized low-density lipoproteins (OxLDLs) into T cells, which induces lipid peroxidation and downstream activation of p38 kinase, thereby impairing antitumor immune responses by restraining the secretion of antitumor factors, including interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α)[13].

Overall, CRC remains a leading cause of cancer-related mortality worldwide[14], as its progression is intricately linked to metabolic reprogramming that sustains rapid proliferation, immune evasion, and therapeutic resistance. According to 2022 cancer statistics from China and the United States, a significant recent decline in the age-standardized incidence and mortality rates of CRC has been observed in the United States[15]. In contrast, China has experienced an increase in the incidence of CRC[15]. Epidemiological data from Western countries indicate a rising incidence of CRC among individuals under 50 years of age[16]. Approximately 75% of newly diagnosed CRC patients present with nonmetastatic, early-stage disease, which is amenable to curative-intent treatment[17]. However, 5%-30% of patients with colon cancer experience disease recurrence[18,19].

Moreover, lipid metabolism has emerged as a critical pathway in CRC, and a better general understanding of lipid metabolic reprogramming in CRC may result in new prognostic biomarkers and therapeutic targets for cancer treatment. However, despite advances in understanding lipid-driven oncogenesis, critical gaps remain. The context-dependent roles of lipid species, compensatory metabolic pathways under therapeutic pressure, and the spatiotemporal heterogeneity of lipid metabolism in CRC subtypes require further exploration. This review synthesizes the current knowledge of lipid metabolism in CRC, emphasizes its role in tumor initiation, immune modulation, and therapeutic resistance, and proposes strategies to exploit these pathways for precision oncology.

HIGH-FAT DIETS AND CRC

High-fat diet and CRC progression

Accumulating evidence indicates that certain types of cancer, including CRC, develop as a result of a high-fat diet (HFD) or are associated with an HFD, and that the disease may progress in a manner that involves signaling pathways that modulate lipid metabolism. Intriguingly, obesity is also an independent risk factor for the development of distant metastasis, therapy resistance, and death in CRC patients.

HFD and CRC carcinogenesis

A HFD influences CRC tumorigenesis, which is supported by the observation that a Western diet is associated with an increased risk of CRC in humans[20]. An HFD induces significant alterations in the gut microbiota composition and fosters a tumor-promoting environment. In murine models, HFD-fed mice present enriched pathogenic genera, including Marseille-P5997 and Alistipes spp. 5CPEGH6, and depleted beneficial species, such as Parabacteroides distasonis, along with impaired gut barrier function[21]. These microbiota shifts were shown to directly promote colorectal tumorigenesis in germ-free mice, which confirms the essential role of gut microbes in HFD-associated CRC[21]. Specific HFD-enriched bacterial strains, such as Coriobacteriaceae Cori. ST1911, has been shown to activate the CPT1A-extracellular regulated protein kinases (ERK) axis, which increases acyl-carnitine levels and fuels CRC progression[22]. Conversely, probiotic interventions, such as Lactobacillus La.mu730, can antagonize pathogenic bacterial colonization and restore barrier integrity, suggesting therapeutic potential. A HFD was also shown to alter bile acid metabolism, which increases nonclassical amino acid-conjugated bile acids, including cholic acid derivatives that stimulate intestinal stem cell growth, and of these microbes, Ileibacterium valens and Ruminococcus gnavus were identified as key synthesizers[23]. An HFD also promotes the generation and functional enhancement of LGR5 + intestinal stem cells[24]. This activity increases the oncogenic capacity of peroxisome proliferator-activated receptor (PPAR), including PPAR-δ, together with other PPAR isotypes (PPAR-α and PPAR-γ, which are predominantly present in liver and adipose tissue, respectively), which increases FAO, upregulates PPAR target genes and alters the intestinal stem cell niche[24,25]. These findings indicate that early excessive fat intake may predispose individuals to primary CRC tumorigenesis (Figure 1).

Figure 1 Schematic overview of high-fat diet-induced pathological remodeling in the intestinal microenvironment. High-fat foods promote microbial dysbiosis and metabolic reprogramming, leading to stem cell expansion, lipid accumulation, and immunosuppression. LPC: Lysophosphatidylcholine; LPA: Lysophosphatidic acid; PPAR: Peroxisome proliferator-activated receptor; EMT: Epithelial-mesenchymal transition; IFN: Interferon; TNF: Tumor necrosis factor; PD-1: Programmed cell death protein 1; CD: Cluster of differentiation; SREBP: Sterol regulatory element-binding protein; LDL: Low-density lipoprotein; MAPK: Mitogen-activated protein kinase.

HFD and CRC metastasis

Although increased fat uptake seems to be associated mostly with primary tumor initiation and growth, metastasis has also been linked to an HFD. For example, in prostate cancer, a HFD activates the sterol regulatory element-binding protein (SREBP)-dependent lipogenic program, promotes lipid accumulation, and alters lipid metabolism and the mitogen-activated protein kinase (MAPK) signaling pathway, which leads to increased invasiveness and metastatic potential in prostate cancer cells[26]. Similarly, a HFD causes an increase in low-density lipoprotein (LDL) levels, which correlate with increased liver metastases in CRC patients[27]. LDL enhances CRC cell migration and stemness by upregulating genes such as Sox2, Oct4, and Bmi1 while activating reactive oxygen species (ROS)-mediated MAPK signaling, a key driver of metastatic progression[27]. Genetic interactions, such as a HFD combined with HNF1A mutations, amplify Wnt/β-catenin signaling, which fosters polyp formation and potential metastatic transformation[28]. Obesity induced by a HFD disrupts CD4 + T-cell function in the tumor immune microenvironment, reduces IFN-γ and TNF-α production and increases programmed cell death protein 1 (PD-1) expression. This immunosuppressive milieu weakens antitumor responses and accelerates CRC progression and metastasis[29].

Additionally, the role of PPAR in cancer has been increasingly discussed. To date, three PPARs have been identified and are referred to as PPAR-α, PPAR-δ/β and PPAR-γ. Among them, PPAR-δ is more broadly expressed, and its expression level is much higher in the gastrointestinal system than in other tissues. Similarly, a report investigating the expression of PPAR-δ revealed that this protein accumulates in human CRC cells. As mentioned previously, an HFD promotes tumorigenesis by increasing the number of cells per crypt, increasing proliferation, and enhancing the regenerative capacity of intestinal progenitor stem cells (LGR5 +), in which PPAR-δ is activated, and ultimately increasing FAO[24]. Moreover, PPAR-δ expression in primary tumors was found to be associated with significantly shorter metastasis-free survival in a cohort of patients with stage III CRC. Mechanistically, further investigation revealed that PPAR-δ expression in CRC cells strongly promotes epithelial-mesenchymal transition (EMT). Moreover, abnormal metabolites, including lysophosphatidylcholine and its downstream metabolite lysophosphatidic acid (LPA), were identified as the most significantly upregulated outlier metabolites in HFD-fed mice. The oncogenic role of LPA was further confirmed by its ability to induce CRC cell proliferation, accelerate cell cycle progression, and disrupt intercellular tight junctions[21].

Intriguingly, the interval of HFD intervention has controversial effects on CRC metastasis. A study conducted by Xiang et al[30] revealed that a short-term HFD may decrease the risk of CRC metastasis by promoting the activation of adipose tissue macrophages and increasing the recruitment of CD4 + and CD8 + T cells to visceral adipose tissue. This inhibits metastatic seeding, which implies that an HFD exerts a protumorigenic effect only when obesity is induced by a chronic HFD.

Generally, an HFD has considerable potential for CRC initiation, growth and metastasis formation, as it influences the composition of the gut microbiota, increases the number of intestinal stem cells as well as their proliferation and function, and activates associated transcription pathways. These findings suggest that controlling obesity may help in the treatment of CRC (Figure 1).

EFFECT OF LIPIDS IN CRC

FAs

FAs, characterized by a carboxylic acid group and a hydrocarbon chain of varying carbon chain lengths and degrees of unsaturation (i.e., number of double bonds), serve as the primary building blocks of several specific lipid classes, including phospholipids, sphingolipids, and triglycerides[31]. During oncogenic transformation and progression, cancer cells undergo metabolic reprogramming and prominently exhibit a preference for aerobic glycolysis over oxidative phosphorylation for energy production-a phenomenon known as the Warburg effect[32]. In this metabolic alteration, increased de novo biosynthesis and exogenous FA uptake by cancer cells sustain their proliferation and provide an essential energy source during conditions of metabolic stress[7]. To delve deeper into these lipid metabolic alterations, the following section provides an in-depth analysis (Figures 2 and 3).

Figure 2 The process of lipid metabolism: Lipid metabolic processes include de novo lipogenesis, fatty acid uptake, fatty acid oxidation, cholesterol synthesis and lipid storage. Tumor cells enhance lipid metabolism by augmenting exogenous lipid uptake and de novo lipogenesis, thereby elevating intracellular lipid content. Upregulation of lipid transport proteins-including cluster of differentiation 36, fatty acid transport proteins, fatty acid-binding proteins, and low-density lipoprotein receptor promotes the cellular internalization of saturated fatty acids (SFAs), polyunsaturated fatty acids, and cholesterol. Meanwhile, endogenous lipid synthesis originates from mitochondrial citrate exported to the cytosol via the tricarboxylic acid cycle citrate carrier (SLC25A1). This citrate is converted to acetyl-CoA by adenosine triphosphate (ATP)-citrate lyase, serving as a substrate for sterol regulatory element-binding protein-driven synthesis of SFAs and cholesterol. The process is catalyzed by key enzymes such as fatty acid synthase and stearoyl-CoA desaturase, which introduce double bonds into SFAs to generate monounsaturated fatty acids. Subsequently, fatty acids are activated by acyl-CoA synthetases to form acyl-CoA, which fuels either phospholipid synthesis (e.g., phosphatidylcholine via glycerol-3-phosphate acyltransferase/acylglycerol-3-phosphate acyltransferase enzymes) or fatty acid oxidation in mitochondria to generate ATP and support tumor growth. CD: Cluster of differentiation; FATPs: Fatty acid transport proteins; FABPs: Fatty acid-binding protein; LDLR: Low-density lipoprotein receptor; ACSLs: Acyl-CoA synthetase long chain; CPT: Carnitine palmitoyl transferase; FAO: Fatty acid oxidation; TCA: Tricarboxylic acid cycle; GLS: Glutaminase; GLUD: Glutamate dehydrogenase; IDH: Isocitrate dehydrogenase; ACLY: Adenosine triphosphate-citrate lyase; CE: Cholesterol ester; HMGCS: Hydroxy methyl glutaryl-CoA synthase; HMGCR: 3-hydroxy-3-methylglutaryl-CoA reductase; HMG: Hydroxymethylglutarate; FADS2: Fatty acid desaturase 2; ELOVLs: Elongation of very long-chain fatty acid protein; ACC: Acetyl-CoA carboxylase; FASN: Fatty acid synthase; SCD: Stearoyl-CoA desaturase; MUFAs: Monounsaturated fatty acids; PUFAs: Polyunsaturated fatty acids; GPAT: Glycerol-3-phosphate acyltransferase; LPA: Lysophosphatidic acid; DAG: Diacylglycerol; DGAT: Diacylglycerol acyltransferase; ATGL: Adipose triglyceride lipase; TAG: Triacylglycerol; PC: Phosphatidylcholine; PCho: Phosphorylcholine; ChoK: Choline kinase; GPC: Choline glycerophosphate.

Figure 3 Lipid metabolic alterations in colorectal cancer cells. Hypoxia and signal pathway including transforming growth factor-β1 and phosphatidylinositol 3-kinase/protein kinase B attributes to the reprogramming lipid metabolism in colorectal cancer cells. CD: Cluster of differentiation; FATPs: Fatty acid transport proteins; FABPs: Fatty acid-binding protein; LDLR: Low-density lipoprotein receptor; DAG: Diacylglycerol; DGAT: Diacylglycerol acyltransferase; LD: Lipid droplet; PI3K: Phosphatidylinositol 3-kinase; AKT: Protein kinase B; TGF: Transforming growth factor; TAG: Triacylglycerol; SREBP: Sterol regulatory element-binding protein; PPAR: Peroxisome proliferator-activated receptor; HIF: Hypoxia inducible factor; FASN: Fatty acid synthase; ACSL: Acyl-CoA synthetase long chain; HMGCR: 3-hydroxy-3-methylglutaryl-CoA reductase.

De novo biosynthesis of FAs

The de novo FA biosynthesis pathway is a conserved metabolic process in which carbon atoms derived from carbohydrates (e.g., glucose) and specific amino acids (e.g., glutamine) are enzymatically converted into FAs. Cancer cells frequently exhibit hyperactivation of this anabolic pathway even in the presence of exogenous lipid sources, a phenomenon driven by metabolic reprogramming. Acetyl-CoA serves as the primary substrate for FA synthesis and is primarily sourced from citrate by ATP-citrate lyase (ACLY) or acetate (catalyzed by acetyl-CoA synthetase, ACSS2)[7]. Glucose-derived citrate, produced through glycolysis and the tricarboxylic acid cycle, is processed by ACLY into acetyl-CoA[33,34]. Acetyl-CoA is then carboxylated by ACC1 to form malonyl-CoA[35], which is subsequently utilized by FASN to synthesize palmitate (C16:0), the final product of de novo lipogenesis[36]. Notably, CRC cells persistently activate this anabolic pathway even under lipid-replete conditions, which underscores their unique reliance on endogenous lipid synthesis.

In addition to its metabolic role, ACLY drives CRC metastasis through noncanonical signaling mechanisms. Studies have revealed that ACLY directly stabilizes β-catenin (CTNNB1) via physical interaction and amplifies Wnt/β-catenin signaling to activate EMT-related genes, which increases CRC cell migration and invasiveness (supported by preclinical models and findings in patient tissue)[37]. Intriguingly, ACLY expression is positively correlated with CTNNB1 levels, and their colocalization is predominant at invasive fronts in tumors. Furthermore, the ACLY/SREBP-1 axis regulates the sensitivity of CRC cells to apoptosis: The knockdown of ACLY or SREBP-1 significantly sensitizes cells to chemotherapy-induced apoptosis by suppressing 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), LDL receptor (LDLR), and ACC1 expression, which highlights its potential as a therapeutic cotarget[38]. In addition, FASN overexpression is associated with advanced stages of CRC and CRC metastasis[39]. FASN has been found to promote angiogenesis in CRC. In one study, FASN knockdown decreased microvessel density in CRC cells and resulted in ‘normalization’ of the tumor vasculature. Further investigation revealed that mechanisms associated with antiangiogenic effects included downregulation of vascular endothelial growth factor (VEGF)¹⁸⁹, upregulation of the antiangiogenic isoform VEGF^165b and a decrease in the expression and activity of matrix metalloproteinase-9[40].

SREBPs, particularly SREBP-1, directly activate FASN to drive de novo lipogenesis. In CRC tissues, SREBP-1 colocalizes with FASN and the proliferation marker Ki-67, mechanistically coupling lipid synthesis to cell cycle progression[41]. Paradoxically, pharmacological FASN inhibition (e.g., cerulenin) triggers compensatory upregulation of SREBP-1 and FASN by a transcriptional feedback loop, which represents an adaptive survival mechanism in lipid-deprived tumors[41]. This resilience underscores the necessity for dual-targeting strategies against both lipogenic enzymes and their upstream regulators. In addition to its role in lipogenesis, SREBP-1 promotes CRC aggressiveness through tumor-promoting signaling pathways. SREBP-1 induces matrix metalloproteinase-7 expression via nuclear factor kappa-B (NF-κB) activation, enhancing tumor cell invasiveness and angiogenesis[42]. SREBP-1 also functions jointly with the protein kinase B (AKT)/adenosine 5’-monophosphate-activated protein kinase (AMPK) signaling axis to upregulate glycolytic and lipogenic enzymes, thereby supporting metabolic flexibility in cancer stem cells (CSC)[42]. Similarly, SREBP-2 silencing disrupts mitochondrial respiration and FAO, which impairs tumor sphere formation and xenograft growth[43]. SREBP activity is further modulated by diverse metabolic and signaling inputs. Notably, in CRC cells, polyunsaturated FAs (PUFAs), such as docosahexaenoic acid and oleic acid, activate SREBP-2 independently of cholesterol levels or endoplasmic reticulum (ER) stress, which suggests alternative lipid-sensing mechanisms[44]. Conversely, dietary and pharmacological interventions can suppress SREBP signaling. Metformin, for example, alters the gut microbiota to increase the levels of short-chain FAs (SCFAs), such as propionate and butyrate, which inhibit SREBP target genes and reduce the tumor burden in obesity-associated CRC models[45]. Additionally, the SREBP-cleavage-activating protein (SCAP)/SREBP-1 pathway is a critical upstream node; berberine inhibits lipogenesis by downregulating SCAP, thereby blocking SREBP-1 maturation and its nuclear translocation[46]. Transcriptional coactivators such as peroxisome proliferator-activated receptor-gamma coactivator (PGC)-1α also regulate SREBP-1 indirectly by upregulating FASN expression, which increases CRC cell proliferation and tumorigenicity[47]. Furthermore, interleukin (IL)-6 and other cytokines in the TME may interfere with SREBP-1-mediated lipogenesis, linking inflammation to metabolic reprogramming in CRC[48].

Subsequent desaturation of palmitate by stearoyl-CoA desaturase (SCD) produces monounsaturated FAs (MUFAs), whereas elongation by ELOVL6 adds two-carbon units to palmitate to form the saturated FA stearate. SCD-mediated desaturation represents a critical step in cell survival, as saturated FA accumulation induces lipotoxicity and ER stress[49]. SCD1 specifically catalyzes the synthesis of MUFAs from ∆9-saturated FAs. PUFAs can then be generated through further desaturation of MUFAs by FA desaturases 1 or 2 (FADS1 or FADS2)[50]. Moreover, increased SCD1 expression has been reported in CRC tissues[51]. An increase in SCD1 is related to the accumulation of MUFAs, increased EMT, and the inhibition of PTEN, which promotes CRC metastasis[52]. Although SCD inhibition may be an attractive therapeutic target, pharmacological inhibitors have demonstrated only modest efficacy in cancer treatment[53]. This limited response suggests that cancers resistant to SCD inhibition may utilize alternative desaturation pathways to generate functionally essential lipid species. Indeed, FADS2 plays a dominant role in FA desaturation within cancers and primary tumors that are resistant to SCD inhibitors[54]. Furthermore, during proliferation, these resistant cells utilize FADS2-dependent desaturation of palmitate to sapienate (16:1n-10) to support membrane synthesis[52].

PUFAs can be divided into ω-3 PUFAs and ω-6 PUFAs. It has been reported that, compared with normal mucosa, CRC tissues contain significantly greater levels of PUFAs, including both ω-3 PUFAs and certain ω-6 PUFAs[55], which is consistent with a recently released lipidomic study of CRC tissues[56]. Intriguingly, in vitro studies have revealed a paradoxical depletion of PUFA levels in CRC cell culture media over time, which suggests the active utilization of these lipids by cancer cells to meet their metabolic demands[57].

Arachidonic acid (AA) is synthesized from linoleic acid through sequential catalysis by FADS2 (Δ6-desaturase), ELOVL5 (elongase), and FADS1 (Δ5-desaturase), and of these, FADS2 serves as the rate-limiting enzyme[58]. The correlation between FADS2 expression and cancer has been described. Along with increased FDAS2 activity, the levels of metabolites from AA, including prostaglandin E2 (PGE2), are reportedly increased in cancerous tissue compared with adjacent normal tissue[59], which suggests that FADS2-mediated AA synthesis promotes tumorigenesis in a PGE2-dependent manner. PGE2, one of the most abundant prostanoids in CRC, is synthesized through the catalytic actions of cyclooxygenases (COX) (COX-1 and COX-2), which oxidize AA derived from membrane phospholipids. Consistent with observations in breast tumors, both PGE2 Levels and COX-2 expression are significantly elevated in human colon tumors compared with adjacent normal tissues. Notably, cyclooxygenase inhibitors (e.g., aspirin and other nonsteroidal anti-inflammatory drugs) suppress PGE2 production and exert anticancer effects, as demonstrated by long-term randomized controlled trials that have shown that regular aspirin use for > 20 years reduces CRC incidence and mortality. Mechanistically, PGE2 activates four G protein-coupled receptors (EP1-EP4), which mobilizes intracellular calcium ion and modulates adenylyl cyclase activity, thereby driving oncogenic signaling via the phosphoinositide 3-kinase (PI3K)-AKT, MAPK, and β-catenin pathways[60]. The conserved role of the FADS2-AA-PGE2 axis across cancers highlights its potential as a pan cancer therapeutic target.

ELOVL5, which is critical for the synthesis of very long-chain PUFAs, is overexpressed in CRC tumors and is correlated with poor survival[61-63]. Epigenetic evidence has shown that DNA hypermethylation of the ELOVL5 promoter in CRC cell lines suppresses its expression and that this methylation signature correlates with improved survival in the cancer genome atlas cohorts, which implicates ELOVL5 downregulation as a protective factor[64].

In conclusion, the FA metabolic network shapes the invasive and drug-resistant phenotypes of CRC through the regulation of enzyme activity, interactions with transcription factors, and crosstalk with the TME.

Exogenous uptake of FAs

Recent studies have reported that tumors also take in FAs from the tumor environment, which suggests that FA uptake may be as important for tumor progression as de novo synthesis. The FA translocase SCD1 facilitates FA uptake in CRC cells, a process exploited by tumors to fuel metabolic demand. Studies have revealed that CD36 overexpression enhances FA uptake and promotes metastasis in CRC. Mechanistically, the upregulation of CD36 activates MMP28, a metalloproteinase that degrades E-cadherin, thereby promoting EMT and metastatic dissemination[65]. Additionally, acidic conditions in the tumor niche activate transforming growth factor (TGF)-β2 signaling, which promotes CD36 translocation via protein kinase C-ζ and enables FA uptake and storage in lipid droplets (LDs). These LDs serve as energy reservoirs that support anoikis resistance and metastasis[66]. The interplay between the CD36 and PPAR signaling pathways also modulates CRC progression. For example, the long noncoding RNA TINCR/microRNA-107 axis regulates CD36 expression, suppresses PPAR-mediated lipid metabolism and inhibits CRC proliferation and apoptosis[67].

Other transporters, including FA-binding proteins (FABPs) and FA transport proteins (FATPs), are also implicated in CRC pathogenesis. FABP dysregulation has been increasingly implicated in the pathogenesis of CRC, as emerging evidence highlights the potential of FABPs as diagnostic biomarkers and therapeutic targets. Aberrant DNA methylation-mediated regulation of FABP5 (epidermal-type FABP) drives CRC malignancy. Compared with that in adjacent normal tissue, hypomethylation of the FABP5 promoter is correlated with its upregulation in metastatic CRC cells and tumor tissues[68]. This hypomethylation is linked to splice variants of DNA methyltransferase 3B, which suggests that epigenetic dysregulation is a key mechanism in CRC. Furthermore, FABP5 forms a feed-forward loop with NF-κB: NF-κB promotes FABP5 transcription, whereas FABP5 increases NF-κB activity via IL-8 production, which fosters inflammation and tumor progression[68]. This loop underscores the dual role of FABP5 in lipid metabolism and proinflammatory signaling in CRC.

FABP6 and FABP1 are overexpressed in CRC tissues compared with normal colonic mucosa[69]. Further enrichment analysis revealed that overexpression of FABP6 is correlated with poor overall survival of CRC patients and that FABP6 regulates microRNA-mediated cell proliferation via the insulin-like growth factor (IGF) signaling pathway[69]. Moreover, FABP6 contributes to oxaliplatin resistance in CRC. Kruppel-like factor 5 directly binds to the FABP6 promoter to increase its transcription[70]. Elevated FABP6 expression promotes LD accumulation, proliferation, and chemoresistance in CRC cells, as demonstrated in vitro and in nude mouse models[70]. FABP4 (adipocyte FABP or aP2) is elevated in the plasma of CRC patients who are obese and is linked to EMT[71]. The FABP4 protein is clinically associated with the tumor node metastasis stage, differentiation, and lymph node metastasis in CRC[71].

Studies have demonstrated that SLC27A2 is significantly overexpressed in CRC tissues and is closely associated with PPARs, especially PPARG, which is also highly expressed in CRC[72]. Mechanistically, SLC27A2 facilitates FA uptake and β-oxidation by modulating PPARs through nongenomic cross-pathway interactions, which highlights its central role in CRC metabolic adaptation. Another key downstream target of SLC27A2 is ATP-binding cassette subfamily D member 3 (ABCD3/PMP70), the most abundant peroxisomal membrane protein. SLC27A2 overexpression in CRC disrupts ABCD3 function, which potentially alters FA metabolism and contributes to tumor progression[72]. In summary, exogenous FA uptake plays a significant role in the progression of CRC and involves various transporters and complex regulatory mechanisms. These findings offer new potential targets for the diagnosis and treatment of CRC.

FAO

A critical question concerning the role of FAO in CRC pathogenicity is whether FAO enzymes or their regulators are dysregulated. Studies have revealed the overexpression of various FAO enzymes, including CPT1A[73], CPT1B[74], CPT1C[75], carnitine palmitoyl transferase (CPT) 2[76], the carnitine transporter CT2[77], CD36[78] and acyl-CoA synthetase long chain 3 (ACSL3)[79], in multiple malignancies. The rate-limiting step of FAO is mediated by CPTs, which shuttle FAs into mitochondria. CPT1C, which is overexpressed in CRC, is transcriptionally activated by hypoxia inducible factor-1α under hypoxic conditions[75]. CPT1C knockdown reduces FAO rates, suppresses proliferation, and inhibits metastasis, whereas its overexpression is correlated with poor recurrence-free survival. However, hypoxia inducible factor-1α inhibited by oroxylin A paradoxically increases FAO while suppressing tumor growth[80], which highlights complex regulatory crosstalk. Similarly, CPT1A is critical in CRC cells exposed to adipocyte-rich microenvironments. CPT1A-dependent FAO stabilizes β-catenin by acetylation, which activates Wnt/β-catenin signaling and the expression of CSC-related genes[81].

In addition, once they migrate into blood vessels, the first challenge to the survival of cancer cells is detachment-triggered metabolic stress. When detached from the extracellular matrix, normal and nonmetastatic cells undergo anoikis, a type of caspase-mediated apoptosis[82,83]. This detachment, which results in ATP deficiency is associated with a decrease in nicotinamide adenine dinucleotide phosphate (NADPH) and an increase in ROS[82]. In CRC cells, CPT1A mitigates ROS accumulation, enabling anoikis resistance and lung colonization[84]. Consistently, ACSL3, an enzyme that converts free FAs into acyl-CoA esters, is upregulated by TGF-β1 via the SREBP1 pathway, which enables ATP production and NADPH reduction to fuel EMT, thus promoting CRC metastasis[79]. Moreover, nuclear valosin-containing protein binds to histone deacetylase 1 and facilitates its degradation, thus promoting the transcription of FAO genes, including CPT1A[85]. Additionally, RNF183 increases FAO by upregulating enzymes such as CPT1A, which promotes the acquisition of CSC properties[86]. Paradoxically, peritoneal metastases exhibit reduced CPT1A expression[87], suggesting context-dependent FAO roles.

Phospholipids

Phospholipids, a class of lipids containing phosphoric acid groups, are fundamental structural components of biological membranes, where they play pivotal roles in maintaining membrane integrity, dynamic fluidity, and cellular signal transduction. These amphipathic molecules can be classified into two major subclasses, phosphoglycerides (glycerophospholipids) and sphingolipids, both of which contribute to the asymmetric organization of lipid bilayers and the formation of membrane microdomains (e.g., lipid rafts). In addition to their structural functions, phospholipids have been increasingly recognized as critical regulators of oncogenic processes, including tumor cell proliferation, migration, and metastasis, through the modulation of membrane receptor localization, second messenger generation (e.g., phosphatidylinositol-4,5-bisphosphate), and extracellular vesicle (EV)-mediated communication.

Phosphoglycerides

Phosphoglycerides, a major class of membrane lipids, are classified into four principal categories: Phosphatidylcholine (PC), phosphatidylethanolamine, phosphatidylinositol (PI), and phosphatidylserine (PS). As phospholipids are the major component of the cell membrane, changes in phospholipid composition, such as increased total phospholipid content and altered charge distribution, may increase cell membrane fluidity and signal transduction capacity, thereby promoting invasion and metastasis[88]. Phosphoglycerides and their derivatives are important for maintaining some malignant cancer phenotypes and have been linked to poor cancer prognosis. PC accounts for 40%-50% of all phospholipids in eukaryotic membranes and plays important roles in cellular structure and biological processes[89]. Emerging evidence suggests that PC may serve as a clinically relevant biomarker in CRC. Choline kinase α (CKα), the enzyme responsible for PC biosynthesis from choline, is frequently overexpressed in CRC tissues. This elevated CKα expression not only drives PC accumulation but also promotes CRC progression[90]. Notably, while CKα overexpression is strongly associated with metastatic potential, it is not significantly correlated with primary tumor characteristics, including size, histological grade, or local invasion depth. Importantly, high CKα protein expression serves as an independent prognostic indicator that is correlated with reduced overall survival in both early- and advanced-stage CRC patient cohorts[90]. Platelet-activating factors (PAFs) derived from plasmanyl-PCs exhibit context-dependent effects in CRC[91]. In one study involving APC (Min/+) mice, genetic ablation of phospholipase A2 group 7, which metabolizes PAF, significantly reduced colon polyp formation, which suggests that PAF exhibits antitumor activity in the context of APC mutations. Mechanistically, PAF analogs induce apoptosis in colon epithelial cells by dephosphorylating AKT at Ser473 via the β-arrestin 1/PHLPP2 complex and by activating the intrinsic apoptosis pathway[91].

PIs and related PI phosphates are important in cancer-associated signaling pathways. PI3K is an intracellular lipid kinase involved in the regulation of cell proliferation, differentiation and survival. PI3K catalyzes the conversion of PI bisphosphate to PI triphosphate (PIP3)[92]. PIP3 subsequently recruits AKT and activates the mammalian target of rapamycin (mTOR) complex, thus promoting cell proliferation and angiogenesis in colorectal tumors and inhibiting apoptosis and autophagy[93]. Activation of the PI3K/AKT pathway was found to be associated with the prognosis of patients with stage II colon cancer and may be a favorable prognostic factor in colon cancer patients[94].

The alterations in phosphoglyceride metabolism observed in cancer are driven primarily by the dysregulation of key rate-limiting enzymes involved in phosphoglyceride metabolic pathways, and these changes are essential for tumor growth. For example, phospholipase D (PLD) 2 promotes tumor proliferation and invasion by hydrolyzing PC to produce the second messenger phosphatidic acid (PA). Its high expression in CRC is significantly associated with lymph node metastasis, deep invasion, and patient survival[95]. The expression of PS-specific phospholipase A1 (PS-PLA1) is positively correlated with tumor depth and hematogenous metastasis. Patients with high PS-PLA1 expression have poorer disease-free survival, which indicates that the cleavage products of PS may promote tumor invasiveness[96]. In patients with stage II CRC, those with phospholipase A2 (PLA2)-negative tumors have significantly longer disease-free survival times, which suggests that PLA2 may serve as a prognostic predictor for CRC[97]. Interestingly, glycosylphosphatidylinositol-specific PLD (GPI-PLD) regulates carcinoembryonic antigen release from cancer cells, and PA can inhibit carcinoembryonic antigen by inhibiting GPI-PLD activity, which prevents CRC liver metastasis[98].

Moreover, phospholipid-derived signaling molecules, such as LPA, have been identified as potent regulators of CRC cell behavior. LPA has been shown to increase the metastatic potential of CRC cells by promoting their migration, proliferation, and adhesion to extracellular matrix components. This was evident in a study using DLD-1 cells, where LPA stimulated cell migration and adhesion as well as the secretion of angiogenic factors such as VEGF and IL-8[99]. Moreover, LPA functions in diverse biological processes, including cell survival, proliferation, and migration, through a family of G protein-coupled receptors, lysophosphatidic acid receptor (LPAR) 1-LPAR6[100]. In CRC, LPA activates β-catenin via LPAR2 and LPAR3[101]. This involves the phosphorylation of glycogen synthase kinase-3β, the nuclear translocation of β-catenin, and the transcriptional activation of T-cell factor/Lymphoid enhancer-binding factor target genes. LPAR2-mediated β-catenin signaling further synergizes with Kruppel-like factor 5, which increases TCF4 binding and transcriptional activity[101]. LPA also induces RhoA activation and stress fiber formation via LPAR2, which increases cell adhesion, migration, and invasion. Rho-associated kinase (ROCK) and signal transducer and activator of transcription (STAT) 3 cooperate to drive cell cycle progression, and dual inhibition of these pathways blocks the LPA-induced proliferation of HCT116 cells[102]. LPAR2 activates PI3K/AKT and ERK1/2 in response to LPA through interactions with NHERF2, a scaffolding protein critical for CRC cell survival and IL-8 production[103]. Furthermore, NHERF2 deficiency disrupts STAT3 phosphorylation and downregulates the expression of CD24, a glycoprotein linked to cancer stemness and chemoresistance[104]. The STAT3/CD24 axis is essential for NHERF2-mediated tumor growth, as demonstrated by rescue experiments in which CD24 expression was restored in NHERF2-deficient cells. The metabolic rewiring of phosphoglycerides and their derivatives underpins the aggressive progression of CRC as this rewiring dynamically modulates membrane architecture, signaling networks, and cellular plasticity. By dissecting the interplay among phosphoglyceride dynamics, oncogenic signaling, and the TME, this work illuminates novel avenues for precision oncology in CRC, where metabolic reprogramming is both a hallmark and an Achilles’ heel.

Sphingolipids

Ceramide, sphingosine-1-phosphate (S1P) and closely related enzymes are important molecules in sphingolipid metabolism. Ceramide, a bioactive mediator of numerous cellular functions, such as apoptosis and cell cycle regulation, is in turn hydrolyzed by ceramidase into sphingosine, which is subsequently phosphorylated into S1P by sphingosine kinase (SphK)[105]. In CRC metastasis suppression, the RNA-binding protein LARP6 functions as a critical regulator by binding to ZNF267 messenger RNA to increase its stability and translational efficiency. Mechanistically, the ZNF267 transcription factor cooperates with LARP6 to repress sphingomyelin synthase 2 promoter activity, which shifts sphingolipid metabolism toward ceramide accumulation while diminishing sphingomyelin synthesis a form of metabolic reprogramming that impedes CRC cell invasiveness and metastatic dissemination[106].

S1P, a significant bioactive lipid member of the sphingolipid family, is generated via the phosphorylation of sphingosine, which is a product of ceramide modification by ceramidase, and is mediated by SphK[105]. Accumulating evidence indicates that increased expression of SphK1 and production of S1P promote cell growth and enhance cell cycle transition from G1 to S phase, leading to malignant tumorigenesis[107]. These biological processes are driven by the activation of five G protein-coupled receptors (S1PR1-5), a process known as “inside-out signaling” by S1P[108]. Recent work has indicated that the link between inflammation and cancer involves the SphK1/S1P/S1P receptor 1 (S1PR1) axis, which contributes to the NF-κB/IL-6/STAT3 amplification loop[109]. Emerging evidence has demonstrated that SphK1 and its metabolic product S1P play critical roles in CRC progression. Comparative analyses revealed significantly elevated SphK1/S1P expression levels in metastatic vs nonmetastatic colon cancer[110]. Importantly, SphK1 expression is strongly positively correlated with both focal adhesion kinase (FAK) and phosphorylated FAK (p-FAK) levels. Clinically, the co-expression pattern of SphK1/FAK/p-FAK is significantly associated with advanced disease characteristics, including higher histopathological grade, advanced Dukes stage, and an increased frequency of lymph node and distant metastases[110,111]. Moreover, silencing SphK1 reduces S1P levels, increases ceramides, including C18:0-ceramide and C18:1-ceramide, activates caspase-3/9, and induces cancer apoptosis[112]. Additionally, liver acid sphingomyelinase inhibits the progression of colon cancer metastasis to the liver through the downregulation of S1P and subsequent macrophage accumulation and TIMP1 production from hepatic myofibroblasts[113]. A series of experimental treatments that target SphK1/S1P have shown potential therapeutic effects in CRC[114,115], which suggests that SphK1/S1P may play a therapeutic role.

Cholesterol

Emerging evidence reveals that cholesterol metabolic reprogramming constitutes a hallmark of malignant transformation, with dysregulated biosynthesis, transport, and efflux pathways observed in tumors.

Enhanced cholesterol biosynthesis

Increased cholesterol biosynthesis is a hallmark of many cancers. Increasing numbers of studies have shown that SREBP2 and its downstream targets, including de novo cholesterol synthesis enzymes, are significantly upregulated in CRC. Specifically, the de novo cholesterol synthesis pathway is hyperactivated in CRC and is driven by the upregulation of key enzymes such as HMGCR and squalene epoxidase (SQLE). The rate-limiting enzyme of the mevalonate pathway, HMGCR, which is regulated by SREBP2, is overexpressed in CRC[56]. RUNX1, a transcription factor that is overexpressed in aggressive CRC subtypes, directly activates HMGCR, which accelerates cholesterol synthesis and tumor proliferation[116]. SQLE, the rate-limiting enzyme in cholesterol biosynthesis, is overexpressed in CRC tissues and is correlated with poor prognosis[117]. Mechanistically, oncogenic pathways such as c-MYC and AP4 transcriptionally activate SQLE, while p53 Loss or mutation further amplifies cholesterol synthesis by reducing the miR-205-mediated suppression of SQLE[118]. This dysregulation promotes tumor growth and liver metastasis, particularly under the influence of hepatocyte growth factor, which activates the SREBP2-dependent cholesterol synthesis pathway via c-Met/PI3K/AKT/mTOR signaling[119]. Pharmacological inhibition of SQLE (e.g., terbinafine) suppresses CRC proliferation and synergizes with chemotherapy, which highlights its therapeutic potential[119].

Cholesterol biosynthesis also plays a critical role in CSC maintenance. Clustered regularly interspaced short palindromic repeats screening revealed that HMGCR and farnesyl diphosphate synthase are critical for the survival of CSCs in the colon[120]. Genetic or pharmacological inhibition of these enzymes impairs spheroid formation and tumorigenicity by suppressing geranylgeranyl diphosphate-mediated activation of the TGF-β pathway; this downregulates the expression of inhibitors of differentiation proteins, which are key regulators of cancer stemness[120]. Similarly, Clostridium symbiosum, a CRC-associated gut bacterium, increases cholesterol synthesis via branched-chain amino acids, which then activate Sonic Hedgehog signaling to promote CSC proliferation[121]. These findings suggest that a sufficient cholesterol supply for the CSC population may be essential to support CRC progression. Cholesterol biosynthesis is a metabolic linchpin in CRC that drives tumor growth, metastasis, and CSC maintenance. Its dysregulation via SREBP2, oncogenic transcription factors, and microbiota interactions positions this pathway as a multifaceted therapeutic target. Overcoming compensatory adaptations and leveraging combinatorial approaches will be critical for exploiting cholesterol metabolism vulnerabilities in CRC.

Increased exogenous cholesterol uptake

Both cholesterol synthesis and cholesterol uptake contribute to intestinal tumorigenesis. Compared with de novo cholesterol synthesis, increased cholesterol uptake might sometimes be more efficient for cancer cells[122]. CRC cells exhibit increased uptake of LDL via LDLR overexpression, which is linked to advanced tumor stages, metastasis, and stemness. LDL promotes cell migration, spheroid formation, and the upregulation of stemness markers, including Sox2 and Oct4, by activating the ROS-MAPK pathways[27]. In liver metastasis, the EFNB2/EPHB4 axis increases LDLR-mediated cholesterol uptake, which drives metastatic growth through STAT3-dependent transcriptional activation[123]. Clinically, elevated LDL-cholesterol levels are associated with CRC risk in non-statin users, which suggests a role for dietary or circulating cholesterol in tumorigenesis[124]. While exogenous cholesterol uptake is crucial, its interplay with de novo cholesterol synthesis is context dependent. For example, APC/KRAS-mutant CRC cells exhibit elevated proprotein convertase subtilisin/Kexin type 9 (PCSK9) expression, which paradoxically suppresses LDLR-mediated cholesterol uptake while driving de novo synthesis via geranylgeranyl diphosphate accumulation[125]. This suggests that metabolic rewiring favors endogenous pathways in certain genetic contexts.

Accumulated cholesterol derivatives: Cholesteryl esters and oxysterols

Cholesterol esterification, which is mediated by SOAT1/ACAT1, is critical for CRC progression. SOAT1 converts free cholesterol into cholesterol esters (CEs), which are stored in LDs to sustain membrane synthesis and signaling. In APC < sup > Min/+ </sup> mice and colitis-associated CRC models, SOAT1 deficiency reduces the tumor burden by normalizing cholesterol homeostasis, whereas its inhibitor, avasimibe, has therapeutic efficacy[126]. Additionally, ceramide accumulation in CRC activates Toll-like receptor 4 (TLR4)/β-catenin signaling, which upregulates SOAT1 and exacerbates CE-driven tumorigenesis. Another study indicated that TLR4 small interfering RNA inhibits cell proliferation, migration and invasion by suppressing SOAT1 expression[127]. Consistently, lysosomal acid lipase, which hydrolyzes CEs, is upregulated in CRC, which suggests active CE metabolism in CRC cells[128].

Consequently, CEs act as readily accessible reservoirs of cholesterol for cancer cells, enabling them to meet increased metabolic demands. This role also explains the cancer-associated upregulation of relevant enzymes such as acyl-CoA: ACAT1 and lipase, which facilitate rapid interconversion between esterified and free cholesterol.

Oxysterols, which are oxidized cholesterol derivatives, exhibit context-dependent effects. Since oxysterols are known liver X receptor (LXR) ligands that repress SREBP signaling, they are likely to inhibit cell proliferation and reduce de novo cholesterol synthesis. 27-hydroxycholesterol (27-HC) suppresses CRC proliferation by inhibiting AKT activation and cyclin-dependent kinases, independent of the estrogen receptor or LXR pathways[129]. Similarly, 27-HC together with 24(R/S), 25-epoxycholesterol inhibits cancer cell proliferation and migration via the modulation of LXR signaling in gastric cancer[130]. Conversely, 25-HC induces anoikis in CRC spheroids by activating the ROCK/Lin11, Isl-1 and Mec-3 kinase/cofilin axis, which leads to cytoskeletal remodeling and caspase-3 activation[131]. Taken together, these findings underscore the complexity of oxysterol signaling, but further investigation is needed.

LIPID METABOLIC CROSS-TALK BETWEEN CELLS IN THE CRC TME

Tumors employ various mechanisms to evade immune system clearance and suppress anti-tumor immune responses[132]. Metabolic reprogramming is one such mechanism in which the metabolic profiles of tumor and immune cells are altered to satisfy the developmental needs of tumor cells and adapt to the complex TME[132]. Normal cells rely on mitochondrial respiration for energy, but tumor cells rely mainly on glycolysis even under aerobic conditions, which is known as the Warburg effect[133]. However, the vigorous reprogramming of lipid metabolism also plays an important role in tumorigenesis and development[134]. Changes in lipid metabolism in tumor cells are not only driven by their own needs, but they also play a pivotal role in driving tumor progression and reshaping the TME, particularly through its crosstalk with immune components (Figure 4).

Figure 4 Mechanism of lipid metabolic reprogramming in immune cells. Overexpression of cluster of differentiation (CD) 36 on CD8 + tumor-infiltrating lymphocytes (TILs) promotes metabolic transition by activating peroxisome proliferator-activated receptor (PPAR), which results in enhanced fatty acid oxidation (FAO). Lipid metabolism in regulatory T cells is enhanced by CD36-PPAR-β signaling, mammalian target of rapamycin-cholesterol signaling, and sterol regulatory element-binding protein (SREBP)-mediated overexpression of fatty acid synthase (FASN). Tumor-associated macrophages can uptake exosomes containing long-chain fatty acids, SREBPs-mediated overexpression of FASN. These effects together promote M2 polarization. Natural killer (NK) cells uptake lipids and activate PPAR signaling, which impairs NK cells’ function. In dendritic cells, CD36/toll-like receptor (TLR) 2/TLR6 complex internalizes oxidized low-density lipoprotein, activates nuclear factor kappa-B/NLRP3/cyclic guanosine monophosphate-adenosine monophosphate synthase-stimulator of interferon genes signaling and CD8 + TILs. Overexpression of fatty acid transport protein 2 in myeloid-derived suppressor cells mediates arachidonic acid uptake and upregulate prostaglandin E2 synthesis. Cancer-associated fibroblasts upregulate FASN and enhance FAO through increasing carnitine palmitoyl transferase 1, which finally promotes epithelial-mesenchymal transition of colorectal cancer cells. CD: Cluster of differentiation; NK: Natural killer; mTOR: Mammalian target of rapamycin; IFN: Interferon; PPAR: Peroxisome proliferator-activated receptor; FASN: Fatty acid synthase; SREBP: Sterol regulatory element-binding protein; Treg: Regulatory T cell; AA: Arachidonic acid; PGE2: Prostaglandin E2; MDSCs: Myeloid-derived suppressor cells; FATP2: Fatty acid transport protein 2; STAT: Signal transducer and activator of transcription; p-STAT: Phospho-signal transducer and activator of transcription; GM-CSF: Granulocyte-macrophage colony-stimulating factor; XBP1: X-box binding protein 1; EV: Extracellular vesicle; HMGB1: High mobility group box-1 protein; OxLDL: Oxidized low-density lipoproteins; TLR: Toll-like receptor; DCs: Dendritic cells; NF-κB: Nuclear factor kappa-B; cGAS/STING: Cyclic guanosine monophosphate-adenosine monophosphate synthase-stimulator of interferon genes signaling; IL: Interleukin; IFN: Interferon; TNF: Tumor necrosis factor; FAO: Fatty acid oxidation; CPT: Carnitine palmitoyl transferase; LCFA: Long-chain fatty acids; LD: Lipid droplet; TAM: Tumor-associated macrophages; CAF: Cancer-associated fibroblast; EMT: Epithelial-mesenchymal transition; CRC: Colorectal cancer.

CD8 + T cells

CD8 + TILs are key antitumor components of the immune system. However, during cancer progression, the cytotoxic abilities of these cells are compromised as they transition into an exhausted state[13]; this behavior is consistent with the decrease in the levels of some effector molecules, including IFN-γ and TNF-α, which are secreted by tumor-infiltrating CD8 + T cells[135]. In response to hypoxia, glucose deficiency and lipid accumulation in the TME, CD8 + TILs often undergo metabolic transitions. One transition involves the switch from glycolysis to FAO, which maximizes the activity of CD8 + TILs so that their antitumor function is maintained. For example, in CD8 + T cells in MC38 CRC cell line-bearing mouse models, FAO increases with the upregulation of CPT1[136]. In addition, the PPAR agonist bezafibrate upregulates CPT1 to increase FAO, which leads to enhanced antitumor effects during anti-PD-1 therapy in animal models and implies that PPAR signaling activates FAO in CD8 + TILs[136]. These findings suggest that the increase in FAO in CD8 + TILs can enhance antitumor effects.

However, excessively elevated lipid metabolism can lead to lipid peroxidation and ROS accumulation within the cell, which further impairs their antitumor effects. In CRC, CD8 + TILs have been reported to upregulate CD36 expression to facilitate extracellular lipid uptake[13]. Mechanistically, CD36 promotes the uptake of OxLDLs into T cells, which induces lipid peroxidation and downstream activation of p38 kinase, thereby impairing antitumor immune responses[13]. Recent studies have shown that PCSK9, an element that regulates lipid oxidation metabolism by inducing the degradation of LDLR in lysosomes, may be a crucial regulator of cancer immunotherapy[137,138]. One CRC study indicated that a PCSK9-neutralizing antibody could enhance the antitumor effects of a PD-1 inhibitor through downregulation of TGF-β expression and increased CD8 + T-cell infiltration[139]. Moreover, the expression of CD8 + T-cell immune checkpoints is positively associated with increased cholesterol accumulation, which may ultimately lead to CD8 + T-cell exhaustion[140]. Interestingly, in melanoma, ACAT1-deficient CD8 + T cells are better than wild-type CD8 + T cells at controlling melanoma growth and metastasis[141]. The ACAT inhibitor avasimibe, which was previously tested in clinical trials for the treatment of atherosclerosis and exhibits a good human safety profile, can be used to treat melanoma in mice and has been shown to have beneficial antitumor effects[141]. Furthermore, S1PR4 signaling suppresses CD8 + T-cell proliferation and survival via the transcriptional regulation of Pik3ap1 and Lta4h[142]. Genetic deletion of S1PR4 delays CRC progression by increasing intra-tumoral CD8 + T-cell numbers and improving clinical outcomes. In conclusion, optimal levels of lipid metabolism are necessary to maintain the antitumor effects of CD8 + TILs, but accumulation of lipids and ROS may impair these effects (Table 1)[11,13,87,136,143-156].

Table 1 Lipid metabolism in immune cells.

Immune cells	Key metabolic alteration	Mechanisms	Effect on immune cells	Ref.
CD8 + T cells	CPT1 (increase)	Enhanced FAO via PPAR signaling activation	Enhanced antitumor function	Chowdhury et al[136]
	CD36 (increase)	CD36-mediated OxLDL uptake (increase)-lipid peroxidation-p38 activation	Impairing anti-tumor immune responses	Xu et al[13]
Treg cells	CD36 (increase)	Upregulated CD36 activates PPAR-β mitochondrial fitness	Enhanced immunosuppression	Wang et al[11]
	FAS (increase)	SREBP/FASN axis drives lipid synthesis, inhibiting SREBP exhibits a rapid clearance of tumor cells	Enhanced immunosuppression	Lim et al[143]
	Cholesterl synthesis (increase)	mTORC1 signaling cholesterol synthesis (increase)	Enhanced immunosuppression	Zeng et al[144]
TAMs	CD36 (increase)	Lipid uptake (increase)	M2 polarization, suppression of CD8 + T cell function	Yang et al[145]
	FAS (increase)	SREBP1-driven fatty acid synthesis (increase)	Orchestrate tumor-associated immunosuppression	Liu et al[146]
	LD (increase)		Tumor growth enhancement, inhibiting LD leads to tumor growth impairment	Wu et al[147]
DCs	Scavenger receptor A (increase)	Lipids uptake (increase)	Impaired antigen presentation	Herber et al[148]
	OxLDL	CD36/TLR2/TLR6 complex internalizes OxLDL-activate NF-κB/NLRP3/cGAS-STING	Enhance anti-tumor immunity	Ma et al[149]
	FFAR2	FFAR2 Loss-IL-27 (increase)-CD8 + T cell exhaustion	Prevent colorectal carcinogenesis	Lavoie et al[150]
NK cells	Obesity	PPAR-α/δ induced by obesity suppresses mTOR glycolysis-IFN-γ (decrease)	Impaired cytotoxicity	Michelet et al[151]
	MSR1, CD36 and CD68	Lipid accumulation	NK cell dysfunction	Niavarani et al[152]
MDSCs	FATP2 (increase)	FATP2 mediates AA uptake PGE2 synthesis	Enhanced immunosuppression	Veglia et al[153]
	XBP1 (increase)	XBP1 drives sEV-cholesterol release immunosuppression		Yang et al[154]
CAFs	FASN (increase)	CAFs secretes lipids absorbed by CRC cells EMT (increase)	Promotion of tumor migration	Gong et al[155]
	CPT1A (increase)	CPT1A (increase) enhances FAO energy supply	Promotion of peritoneal metastasis	Peng et al[87]
	12(S)-hydroxy eicosatetraenoic acid	Induces the retraction of cancer-associated fibroblasts via MLC2, RHO/ROCK and Ca2+ signaling	Enforced invasiveness of CRC	Stadler et al[156]

CD: Cluster of differentiation; Treg: Regulatory T cell; FAS: Fatty acid synthesis; TAM: Tumor-associated macrophages; DCs: Dendritic cells; CPT: Carnitine palmitoyl transferase; FASN: Fatty acid synthase; SREBP: Sterol regulatory element-binding protein; LD: Lipid droplet; OxLDL: Oxidized low-density lipoproteins; FFAR2: Free fatty acid receptor 2; FAO: Fatty acid oxidation; PPAR: Peroxisome proliferator-activated receptor; mTORC1: Mammalian target of rapamycin complex 1; TLR: Toll-like receptor; NF-κB: Nuclear factor kappa-B; cGAS/STING: Cyclic guanosine monophosphate-adenosine monophosphate synthase-stimulator of interferon genes signaling; IL: Interleukin; NK: Natural killer; MDSCs: Myeloid-derived suppressor cells; CAF: Cancer-associated fibroblast; FATP2: Fatty acid transport protein 2; XBP1: X-box binding protein 1; IFN: Interferon; AA: Arachidonic acid; PGE2: Prostaglandin E2; sEV: Small extracellular vesicle; CRC: Colorectal cancer; EMT: Epithelial-mesenchymal transition; MLC2: Myosin light chain 2; Ca²⁺: Calcium ion.

Treg cells

Tregs are CD4 + T cells that express forkhead box protein P3 (FOXP3), which promotes Treg development, FA absorption, oxidative phosphorylation, and FAO. Enhanced lipid uptake, which is driven by the upregulation of CD36, is a prominent feature of Treg cells in the TME. In MC38 cell-induced colon adenocarcinoma, upregulated CD36 is observed in tumor-infiltrating Tregs[11]. The growth deceleration of engrafted MC38 colon carcinoma cells in Treg^CD36-/- mice has also been reported[11]. Mechanistically, another study indicated that increased lipid absorption mediated by CD36 activates the PPAR-β pathway, which improves mitochondrial fitness and the nicotinamide adenine dinucleotide/nicotinamide adenine dinucleotide ratio in Tregs, thus supporting Treg persistence in the TME[11]. Notably, PCSK9 inhibition, which downregulates CD36, also enhances PD-1 efficacy by reducing Treg infiltration and lipid metabolism reprogramming in CRC[139].

SREBP activity is increased in Tregs within tumors and promotes FASN-dependent FA synthesis[143], and FASN is closely associated with the functional maturation of Tregs. Deletion of FASN in Treg cells inhibits the growth of MC38 cells, which indicates that FA synthesis is a key metabolic pathway through which Treg cells maintain their immunosuppressive function[143]. Moreover, FOXP3^CreScap^fl/fl mice exhibit rapid and complete clearance of tumor cells, suggesting that deficiency of SCAP, a factor required for SREBP activity, inhibits tumor growth. Moreover, AKT/mTOR complex 1 signaling in Tregs promotes cholesterol metabolism, and the mevalonate pathway is particularly important for the coordination of Treg proliferation and the upregulation of cytotoxic T-lymphocyte-associated protein 4 and inducible costimulator to establish functional competency in Tregs[144]. In conclusion, targeting CD36 and SREBP in Treg cells may be promising for enhancing the efficacy of cancer immunotherapy.

DCs

DCs are the primary antigen-presenting cells that activate CD8 + T cells and mediate antitumor immune responses. Lipid accumulation in DCs has been observed in colon cell lines and is caused by increased uptake of extracellular lipids due to the upregulation of scavenger receptor A[148]. In addition, DCs with high lipid contents have a substantially reduced capacity to stimulate T cells when they require antigen processing[148].

Moreover, Arf1-ablated tumor cell-derived factors such as OxLDL and high mobility group box-1 protein bind to the CD36/TLR2/TLR6 receptor complex on the DC surface, which facilitates internalization into Rab7 + endosomes[149]. This process activates the NF-κB, NLRP3 inflammasome, and cyclic guanosine monophosphate-adenosine monophosphate synthase-stimulator of interferon genes pathways, forming a “super signal complex” that drives cytokine production to increase CD8 + T-cell infiltration and cross-priming and stemness[149]. In addition, free FA receptor 2 (FFAR2), a receptor for microbiota-derived SCFAs, has been implicated in CRC pathogenesis[150]. In APC^(min/+) FFAR2 ^(-/-) mice, DCs exhibit altered activation states, increased IL-27 production, and enhanced CD8 + T-cell exhaustion, which are correlated with increased tumor burden[150]; this indicates that the loss of FFAR2 promotes colon tumorigenesis. Further analysis revealed that antibodies against IL-27 and an FFAR2 agonist reduce tumorigenesis in mice and might be developed into a treatment for CRC. The development of promising therapies based on DCs is continuing to increase. For example, phase III clinical trials of monocyte-derived DC-based cancer vaccines in patients with CRC (No. NCT02503150, autologous tumor lysate) are ongoing. The preliminary results of a large trial (No. NCT00045968) on the standard treatment of glioblastoma indicate the clinical safety of these vaccines and the potential increase in survival, which suggests their potential for the treatment of patients with cancer[157,158].

NK cells

NK cells are pivotal effectors in the innate immune surveillance of malignancies, where they mediate direct cytolysis of transformed cells and suppress metastatic dissemination[159]. The defective function of NK cells in cancer contributes greatly to tumor escape and metastatic disease[160-162]. NK cell depletion has been shown to be correlated with the recurrence of CRC liver metastasis[163]. This depletion is induced by an accumulation of lactate in the TME, which causes a reduction in the intracellular potential of hydrogen of hepatic NK cells and leads to mitochondrial dysfunction and apoptosis[163].

Similarly, lipid accumulation impairs NK cell cytotoxicity and tumor control[152]. Notably, obese individuals are reportedly deficient in NK cell numbers, and lipid-bearing NK cells fail to eradicate tumor growth in vivo[151]. Mechanistically, the upregulation of PPAR-α/δ induced by obesity inhibits mTOR-mediated glycolysis as well as the downstream transcription of cytotoxic granules and IFN-γ production, thus blunting NK cell antitumor responses[151]. The inhibition of PPAR-α/δ or the blockage of lipid transport can reverse NK cell metabolic paralysis and ameliorate their cytotoxicity. In addition, NK cells in surgically stressed hosts contain increased amounts of lipids, and this lipid-laden phenotype is associated with NK cell dysfunction in CRC-bearing mice[152]. Consistently, in human studies, an analogous mechanism of surgery-induced FA accumulation occurs in patients with CRC following surgical resection[152]. Further experiments indicated that increased lipid accumulation occurs via the upregulation of MSR1, CD36 and CD68 rather than by de novo lipogenesis[152].

However, cholesterol accumulation in NK cells in HCC promotes membrane lipid raft formation and enhances antitumor effects[164]. These findings suggest the complex role of different lipids in NK cells, which requires more careful research.

TAMs

Macrophages, the most abundant myeloid cells that infiltrate the TME, are endowed with a protumoral M2-like phenotype, which facilitates tumor initiation, progression and metastasis[165]. The TME encompasses diverse signaling molecules capable of modulating lipid metabolism in TAMs. This modulation is characterized by increased lipid uptake, increased FASN, and accelerated FAO[12]. These metabolic alterations drive the polarization of TAMs toward the M2 phenotype, which is associated with protumor activities.

Lipid transport proteins, particularly CD36, play crucial roles in lipid accumulation within TAMs[166]. The scavenger receptor CD36 is upregulated in tumor-infiltrating TAMs, and deletion of CD36 in TAMs attenuates liver metastasis in mice, including metastasis of colon cancer cells to the liver, which suggests the therapeutic potential of targeting CD36 as an immunotherapy for liver metastasis[145]. CD36 enhances the polarization of M2-TAMs to promote liver metastasis. Therefore, CD36 may be used as a marker for the identification of liver metastasis progression. Notably, Treg cells repress CD8 + T-cell-derived IFN-γ to maintain the activity of M2-like TAMs by promoting SREBP1-dependent de novo FA synthesis, thereby orchestrating tumor-associated immunosuppression[146]. Subsequent experiments revealed that SREBP1 pathway inhibition synergizes with anti-PD1 therapy to suppress B16 melanoma tumor growth, which provides a rationale for the combination of the SREBP1 inhibitor fatostatin and anti-PD-1 therapy in cancer treatment[146]. Lipid uptake and synthesis promote LD formation. Actually, according to a recent study, LDs are present specifically in TAMs of patients with CRC but not in benign tissue of patients with CRC[147]. Inhibiting LD formation by inhibiting diglyceride acyltransferase (DGAT) leads to tumor growth impairment by a decrease the proportion of CD206 + TAMs, which may be an effective strategy for tumor treatment[147].

In addition to FAs, TAM-modulated cholesterol metabolism also plays a vital role in CRC progression. Moreover, EVs are membrane-bound vesicles that contain different biomolecules that participate in intercellular signal transmission. For example, EVs derived from TAMs can increase migration and invasion by transferring DOCK7 to CRC cells. Further mechanistic exploration revealed that DOCK7 packaged in TAM-EVs could activate RAC1 in CRC cells and subsequently upregulate ABCA1 expression by AKT and FOXO1 phosphorylation, which ultimately regulates cholesterol metabolism and increases membrane fluidity to regulate CRC cell motility and invasiveness[167]. Moreover, ABCA1 is highly expressed in metastatic CRC and therefore has potential utility as a novel therapeutic target. In summary, lipid accumulation impairs the antitumor effect of TAMs. Therefore, targeting lipid accumulation may lead to new targets for antitumor therapy.

MDSCs

MDSCs are pathologically activated cells that display exceptional immunosuppressive ability[168]. MDSCs undergo lipid reprogramming characterized by increased lipid uptake and FAO as lipids accumulate in tumors[169]. Additionally, single-cell transcriptome analysis of polymorphonuclear-MDSCs from human CRC liver metastases revealed enrichment of ferroptotic pathway genes, suggesting that ferroptosis is a unique and targetable immunosuppressive mechanism of polymorphonuclear-MDSCs in the TME[170,171].

PGE2 derived from tumors induces the nuclear accumulation of p50 NF-κB in MDSCs, which diverts their response to IFN-γ toward nitric oxide (NO)-mediated immunosuppression and reduces TNF-α expression; this finding is consistent with the clinical observation of blood monocytic-MDSCs from CRC patients[172]. Mechanistically, p50 NF-κB promotes the binding of STAT1 to regulatory regions of selected IFN-γ-dependent genes, including inducible NO synthase[172]. Further investigation indicated that inhibition of the PGE2/p50/NO axis prevents MDSC suppressive functions and restores the efficacy of anticancer immunotherapy[172].

In addition, MDSCs in CT26 colon carcinoma models have been shown to contain substantially greater amounts of lipids than those in tumor-free mice[153]. FATP2 overexpression in MDSCs promotes AA uptake and PGE2 synthesis, which are controlled by granulocyte-macrophage colony-stimulating factor through the activation of the transcription factor STAT5[153].

X-box binding protein 1 (XBP1) favors the synthesis and secretion of cholesterol, which activates MDSCs and causes immunosuppression through inositol-requiring enzyme (IRE)-1α/XBP1/small extracellular vesicle-cholesterol signaling[154]. In a colon adenocarcinoma model, the combination of IRE-1α inhibition and immunotherapy almost fully suppresses tumor growth, suggesting that blockade of both the IRE-1α-XBP1 pathway and immune checkpoints is a potential therapy for cancer[154]. By targeting metabolic vulnerabilities in DCs and rewiring their immunoregulatory functions, these advances hold transformative potential for CRC immunotherapy.

CAFs

CAFs constitute a predominant stromal population within the TME, where they orchestrate matrix remodeling and metabolic crosstalk to fuel malignancy[173]. Mechanistically, CAFs accumulate and remodel the extracellular matrix, which enables tumor cells to pass through the TME and interact with other cells through the secretion of cytokines and chemokines[174]. Furthermore, CAFs reprogram lipid metabolism via FAO and LD formation, both of which generate oncometabolites (e.g., ketone bodies, lactate) for the supply of nutrient-rich EVs to sustain tumor growth under nutrient stress[175].

To adapt to the TME, CAFs undergo lipid metabolic reprogramming characterized by increased lipid synthesis. Moreover, FASN, a crucial enzyme in FA synthesis, is significantly increased in CAFs[155]. FAs and phospholipids secreted by CAFs are then absorbed by CRC cells after which they promote CRC cell migration[155]. Similarly, metabolomics studies have revealed that CAF-derived lipids promote CRC peritoneal metastasis by enhancing membrane fluidity and that CRC cells take up lipids and lipid-like metabolites secreted by CAFs[176]. Interestingly, SCD, the rate-limiting enzyme in the biosynthesis of unsaturated FAs, is expressed at low levels in peritoneal metastasis. This may be the reason that lipids from CAFs are absorbed by CRC cells, which could compensate for low SCD expression. Conversely, CRC-secreted 12(S)-hydroxy eicosatetraenoic acid enhances the withdrawal of CAFs; this process is mediated by phospholipase C, inositol-3-phosphate, free intracellular calcium ion, calcium ion-calmodulin-kinase-II, RHO/ROCK and MYLK and leads to the activation of myosin light chain 2 and ultimately increases the invasiveness of CRC[156]. Moreover, FAO can be increased by tumor cells to promote lipid accumulation and energy generation. In CRC, CAFs promote the proliferation, migration, and invasiveness of colon cancer cells via CPT1A upregulation to actively oxidize FAs and induce minimal glycolysis[87].

In summary, CAFs and cancer cells communicate by lipid metabolism reprogramming, including lipid synthesis, exogenous lipid uptake and FAO, which contributes to CRC progression and metastasis.

LIPID METABOLISM AND CRC THERAPY

As described above, specific lipids and their metabolites play important roles in modulating CRC progression and the TME; thus, targeting lipid metabolism has promising therapeutic potential (Table 2). Intriguingly, cancer cells undergo metabolic reprogramming through pathway redundancy, which dynamically shifts to parallel biochemical networks when specific enzymatic routes are therapeutically disrupted. Specifically, SCD inhibition seems to be an attractive target for cancer treatment, yet inhibitors that target these metabolic enzymes have shown only modest effects[53]. Indeed, FADS2 has been shown to play a dominant role in FA desaturation in cancer and primary tumors that are resistant to SCD inhibitors. These findings suggest that cancers resistant to pharmacological SCD inhibition could utilize alternative desaturation pathways to generate functionally useful lipid species. Notably, combination therapies that target lipid metabolism and chemotherapy or targeted therapy can be effective in enhancing antitumor efficacy. Clinical trials investigating the preventive effects of lipid metabolism-related therapies on the development of CRC are underway, the results of which are shown in Table 3.

Table 2 Summary of anticancer drugs targeting lipid metabolism.

Target	Drug	Function	Potential side effects	Ref.
FASN	TVB-3664; TVB-3166	Promote tumor apoptosis	Not observed	Zaytseva et al[188]
ACLY	ETC-1002	Combination with IGF1R inhibitor linsitinib significantly inhibits HOXA13-mediated CRC metastasis	NA	Qiao et al[178]
ACCA	TOFA	Induces the apoptosis of cancer cells	Not toxic in mice	Wang et al[180]
ACSS	Triacsin C	Induces cytochrome c release and subsequent cell death	NA	Mashima et al[200]
SCD	Betulinic acid	Induces rapid cell death of colon cancer stem cells	NA	Potze et al[201]
SREBPs	Betulin	Induces cell cycle arrest, autophagy, and apoptosis of metastatic colorectal cancer cells	Shows an inhibitory effect on the cancer cells with non-toxic concentration towards normal cells	Han et al[202]
CD36	ABT-510; SSO	The combination of ABT-510 and bevacizumab was noted in a variety of tumor types including CRC; Inhibits FA uptake and cell growth in colorectal cancer cells cocultured with CAFs	Safe in clinical trials; NA	Luo et al[203]; Gong et al[155]
CPT1	Etomoxir	Enhances the antitumor effect of cisplatin chemotherapy on HCT116 colon cancer cells	NA	Hernlund et al[194]
HADHA	Ranolazine	Inhibits cancer cell growth, survival, and tumorigenicity in mice	Not toxic in mice	Hossain et al[204]
Cer analogs	C2-Cer; LCL-30	Inhibits proliferation and stimulates apoptosis; Induces cell death by reducing the mitochondrial membrane potential	NA; Not found to be toxic in freshly hepatocytes (selective toxicity)	Ahn and Schroeder[205]; Dindo et al[206]
Acid ceramidase	LCL-521	Decreases viability of colon cells	NA	Tirodkar et al[207]
PLD	FIPI	Decreases cancer cell proliferation and migration promoted by stromal fibroblasts	NA	Majdop et al[208]
SphK1	PF-543; RB005	Inhibits CRC cell proliferation. Inhibits CRC cell growth and proliferation	NA; NA	Ju et al[155]; Shrestha et al[209]
SphK2	ABC294640	Induces CRC cell growth inhibition and apoptosis; inhibits colitis-driven colon carcinogenesis	No systemic toxicity	Xun et al[210]; Chumanevich et al[211]
S1PR modulators	FTY-720	FTY720 induces apoptosis in colorectal cancer cells via PP2A activation. FTY720 exhibits an additive effect with 5-fluorouracil, SN-38, and oxaliplatin in standard chemotherapy in patients with CRC	Cytotoxic effects of FTY720 are markedly reduced in normal colon cells	Cristóbal et al[212]
HMGCR	Statins	Decrease cancer mortality and promote longer survival, according to retrospective clinical analysis	NA	Poynter et al[213]
OSC	R048-8071	Dampens cell proliferation, increases apoptosis and decreases cell migration	NA	Maione et al[214]
ACAT1	Avasimin	Increases apoptosis with no clear cytotoxicity	Safe in mouse models of avasimin treatment	Lee et al[215]
LXR	SR9243	Represses lipogenesis and glycolysis in cancer cells; induces apoptosis	Not toxic	Flaveny et al[216]

FASN: Fatty acid synthase; ACLY: Adenosine triphosphate-citrate lyase; ACCA: Acetyl-CoA carboxylase-A; SCD: Stearoyl-CoA desaturase; SREBP: Sterol regulatory element-binding protein; C2-Cer: C2 ceramide; CD: Cluster of differentiation; CPT: Carnitine palmitoyl transferase; PLD: Phospholipase D; SphK: Sphingosine kinase; SIPR: Sphingosine-1-phosphate receptor; HMGCR: 3-hydroxy-3-methylglutaryl-CoA reductase; OSC: Oxido squalene cyclase; LXR: Liver X receptor; FTY-720: Fingolimod hydrochloride; ETC-1002: Bempedoic acid; TOFA: 5-tetradecyloxy-2-furoic acid; SSO: Sulfo-N-succinimidyloleate; IGF1R: Like growth factor 1 receptor; CRC: Colorectal cancer; FA: Fatty acid; CAF: Cancer-associated fibroblast; PP2A: Protein phosphatase 2A; NA: Not appliable.

Table 3 Clinical trials associated with lipid metabolism for colorectal cancer therapy.

Target	Treatment	Phase	Clinical trials
COX	Sulindac combined with erlotinib	Phase II	NCT01187901
COX	Asprin combined with anti-PD-1 Ab	Phase II	NCT03638297
Inhibit autophagy	Hydroxychloroquine and autophinib	Phase II	NCT02316340
Ferroptosis	Neratinib	Clinical	NCT03457896
COX	Celecoxib combined with irinotecan	Phase I	NCT00084721

COX: Cyclooxygenases; PD-1: Programmed cell death protein 1; NCT: National clinical trials.

Targeting FA metabolism in CRC

Given the critical involvement of FAs in sustaining tumor cell proliferation, survival, and metabolic adaptation, selectively targeting FA metabolism has emerged as a promising therapeutic approach for disrupting oncogenic signaling pathways and exploiting cancer-specific metabolic vulnerabilities. Targeting FAs in CRC can be divided into three general strategies: (1) Blocking de novo FA synthesis; (2) Blocking FA uptake; and (3) Blocking FA oxidation.

Elevated lipid biosynthesis, particularly FA and cholesterol metabolism, has been recognized as a critical metabolic adaptation in cancer cells that supports their malignant progression. Consequently, pharmacological inhibitors targeting rate-limiting enzymes within these biosynthetic pathways are emerging as promising frontiers in cancer therapeutics. ACLY is the first rate-limiting enzyme in FA synthesis and converts citrate to acetyl-CoA. Bempedoic acid (ETC-1002), an ACLY inhibitor, has been approved by the Food and Drug Administration for decreasing LDL-cholesterol levels in atherosclerotic cardiovascular disease and has been applied as a cancer therapy[177]. ETC-1002 alone can effectively inhibit CRC metastasis, while the combination of ETC-1002 and the IGF-1 receptor inhibitor linsitinib significantly suppresses HOXA13-mediated CRC metastasis[178]. Similarly, 5-tetradecyloxy-2-furoic acid (TOFA) blocks FA synthesis by inhibiting ACC, the rate-limiting enzyme of the FA synthesis pathway[179]. Further analysis indicated that TOFA is cytotoxic to colon cancer cells, as this agent induces apoptosis through disruption of FA synthesis[180].

Targeting SCD1, the most critical enzyme in MUFA synthesis, has become a potentially effective approach for cancer therapy. In CRC, SCD1 inhibition efficiently reduces free MUFA levels in colon CSCs, which is accompanied by CSC elimination[181]. Mechanistically, the SCD1 inhibitor-induced reduction in MUFAs blocks Wnt signaling, leading to CSC suppression. Several other strategies to inhibit SCD1 have also demonstrated preclinical efficacy. High concentrations of hydrogen gas (H2) suppress CRC growth by inhibiting the phospho-AKT/SCD1 pathway, which suggests that H2 is a potential SCD1 inhibitor[182]. Similarly, RASAL1, a tumor suppressor, downregulates SCD1 via the LXR-α/SREBP1c pathway, which impairs lipid synthesis and CRC proliferation[183]. Overall, SCD1 is a promising effective target for the selective elimination of CSCs in colon cancer.

FASN, a pivotal enzyme governing de novo lipogenesis, is significantly upregulated in CRC tissues compared with adjacent normal mucosa[184]. The PI3K/AKT and MAPK/ERK1/2 signaling pathways, which promote cell proliferation, survival, and antiapoptotic responses, are implicated in the regulation of FASN expression in cancer[185,186]. In CRC cells, pharmacological inhibition of FASN by emodin induces a time-dependent reduction in PI3K/AKT phosphorylation and concurrent upregulation of ERK1/2 phosphorylation[187]. Notably, combined treatment with emodin and cerulenin (a natural FASN inhibitor) synergistically enhances growth suppression and apoptosis, which indicates that dual blockade of FASN enhances antitumor efficacy. TVB-3664, a selective FASN inhibitor, promotes tumor apoptosis and suppresses tumor progression through the dual modulation of phospho-AMPK and phospho-AKT signaling coupled with the activation of β-catenin[188]. In addition, activation of the AKT and AMPK pathways is associated with resistance to TVB-3664 treatment in patient-derived xenograft models. Therefore, the combination of FASN inhibitors with inhibitors of the AKT or AMPK pathway may be a potential therapeutic strategy for CRC[188]. Furthermore, cotargeting FASN (via TVB-3664) and sphingolipid metabolism [via fingolimod hydrochloride (FTY720) significantly inhibits the proliferation and migration of primary CRC cells[189]. Orlistat, a FASN inhibitor, blocks tumor growth in xenografts by causing cell cycle arrest in G1 phase and triggering apoptosis through caspase-3 activation[190].

In addition to synthesis-focused strategies, emerging evidence implicates CD36, a FA transporter, as a druggable target. Palmitic acid, a compound that induces ER stress and transferrin-dependent ferroptosis, preferentially exerts antitumor effects on CD36-high CRC cells[191]. These findings suggest that the CD36 expression status could stratify patients for ferroptosis-inducing therapies. Furthermore, the long noncoding RNA TINCR/miR-107/CD36 axis regulates the PPAR signaling pathway. TINCR overexpression suppresses CRC progression by competitively binding to miR-107, which upregulates CD36 and promotes apoptosis[67]. PPAR pathway inhibition also decreases the number of Treg cells and upregulates PD-1, which improves the function of CD8 + T cells[192]. Furthermore, JC63.1C, an anti-CD36 monoclonal antibody, was found to resensitize lapatinib-resistant xenograft tumors to human epidermal growth factor receptor 2-targeted therapy, particularly in breast cancer[193]. Collectively, these findings position lipid metabolism rewiring spanning FASN-driven synthesis and CD36-mediated uptake as a dual-axis vulnerability in CRC, and combined targeting offers a roadmap to circumvent resistance and amplify therapeutic efficacy.

FAO, a critical energy metabolism pathway in CRC, relies on CPT1A, the rate-limiting enzyme essential for rapid cancer cell proliferation. Pharmacological inhibition of CPT1A with the widely utilized inhibitor etomoxir has demonstrated potent chemo sensitizing effects. For example, a study showed that etomoxir inhibition of FAO enhances the antitumor effect of cisplatin chemotherapy on HCT116 colon cancer cells[194], which is consistent with the finding that blocking CPT1 with etomoxir significantly enhances the cytotoxicity of Ara-C to drug-resistant leukemia cells[195]. In addition, a more selective CPT1 inhibitor, perhexiline, has been approved for the treatment of heart disease[196]. Pharmacological inhibition of CPT2 using perhexiline disrupts NADPH regeneration and redox homeostasis in CRC cells, resulting in increased ROS generation and increased apoptotic cell death following oxaliplatin treatment. Critically, the combination of oxaliplatin and perhexiline has demonstrated synergistic antitumor efficacy, as they significantly suppress tumor progression in both cell-derived xenograft and patient-derived xenograft models[197]. Other CPT1A inhibitors have also shown promise in preclinical CRC models. 2,6-dihydroxypeperomin B, a covalent inhibitor that binds to the Cys96 residue of CPT1A, disrupts the CPT1A-VDAC1 interaction, impairing mitochondrial function and inducing apoptosis in CRC cells[198]. Another study highlighted a novel PPAR-α antagonist with dual activity against CPT1A that causes a reduction in CRC cell viability[199]. The combination of FAO inhibitors and conventional chemotherapeutics has demonstrated enhanced efficacy across preclinical CRC models, which underscores FAO as a pivotal metabolic vulnerability. These findings support the initiation of clinical trials that will evaluate FAO-targeted combinatorial regimens to overcome chemoresistance and improve the outcomes of CRC patients. The anticancer drugs that target FA metabolism are detailed in Table 2[200-216].

Targeting phospholipid metabolism in CRC

Phospholipids, as fundamental structural components of cellular membranes, play critical roles in orchestrating tumor growth, proliferation, and metastatic dissemination through dynamic membrane remodeling and lipid-mediated signaling. Targeting key signaling nodes and rate-limiting enzymes in phospholipid biosynthesis such as PI3Ks, CKα, and lysophosphatidylcholine acyltransferase (LPCAT) represents a promising therapeutic strategy for CRC.

The PI3K/AKT/mTOR signaling axis, which is hyperactivated in > 40% of CRC cases, drives oncogenic progression via PIP3-mediated membrane recruitment and activation of AKT. This pathway is pharmacologically targetable, with several inhibitors already approved for clinical use. BKM120, an irreversible PI3K-specific inhibitor, inactivates PI3K by covalently modifying the Lys-802 residue involved in the phosphor transfer reaction, which inhibits cancer cell growth and has anticancer effects[217]. MK-2206 may effectively inhibit AKT isoforms and thus has antitumor activity[218]. In addition, thymocyte selection associated with high mobility group box acts as a tumor suppressor by inhibiting mTOR signaling in CRC, and rapamycin (an mTOR inhibitor) alone or combined with a PD-1 inhibitor was shown to be more effective than a PD-1 inhibitor alone in a tumor model[219]. Interestingly, a study showed that mTOR inhibition induces insulin receptor substrate-1 expression and abrogates feedback inhibition of the pathway, which results in AKT activation both in cancer cell lines and in patient tumors treated with the rapamycin derivative RAD001[220]. These findings suggest that combination therapy to prevent mTOR and AKT activation may improve antitumor activity. Overall, the PI3K/AKT/mTOR signaling axis represents a promising therapeutic target in CRC. Further mechanistic investigations and clinical validation of this pathway will be essential for developing effective targeted therapies and optimizing treatment efficacy.

In addition to targeting signaling pathways, key enzymes and phospholipid derivates may also represent potential therapeutic targets. LPCAT3, a phospholipid-remodeling enzyme that catalyzes the incorporation of PUFAs at the sn-2 site of lysophospholipids, gives rise to polyunsaturated phospholipids and has been found to play a vital role in regulating intestinal stemness and tumorigenesis[221]. Inhibition of LPCAT3 disrupts phospholipid remodeling, thereby increasing membrane saturation and stimulating de novo cholesterol biosynthesis, which collectively drives intestinal stem cell hyperproliferation[221]. Pharmacological inhibition of cholesterol synthesis normalizes crypt hyperproliferation in LPCAT3-deficient organoids and mice, which suggests that LPCAT3 could be used as a strategy for therapeutic intervention in CRC. Moreover, LPCAT2, an LD-localized enzyme that supports PC synthesis, drives resistance to cell death after 5-fluorouracil and oxaliplatin treatments both in vitro and in vivo through LD accumulation. Mechanistically, LD accumulation impairs caspase cascade activation and ER stress responses[222]. Additionally, choline kinase alpha (ChoKα), the first enzyme in the Kennedy pathway, is responsible for the synthesis of the major phospholipid of the plasma membrane, PC. In addition, ChoKa is overexpressed in various tumors[223-225]. The ChoKα-specific inhibitors MN58b and TCD-717 have demonstrated potent antitumoral activity both in vitro and in vivo against CRC-derived cell line xenografts, especially when combined with 5-fluorouracil; therefore, this combination may be a new alternative for the treatment of colon tumors[226]. Similarly, other enzymes, including PLD, have been investigated in CRC. For example, in one study, PLD1 inhibition by a drug discovered by computational modeling subsequently increased the phagocytosis of cancer cells by macrophages through the surface expression of costimulatory molecules; as a result, cancer cells became more susceptible to cytotoxic T-cell-mediated killing[227]. Furthermore, the combination of PLD1 inhibitors and anti-programmed cell death ligand 1 antibodies synergistically enhances tumor suppression. Notably, COX inhibitors have shown efficacy in gastrointestinal cancer[228]. Neoadjuvant PD-1 blockade with toripalimab and celecoxib has shown efficacy in locally advanced CRC[229]. Given the role of LPA and LPA receptors in CRC, effective targeting of these receptors has become an important goal in tumor research. For example, cyclic PA, a structural LPA analog, induces G0/G1 arrest in DLD-1 cells by downregulating cyclin D1 and AKT phosphorylation[230]. The inhibition of autotaxin, the enzyme that generates LPA, synergizes with EZH2 inhibition to suppress CRC growth[231]. LPAR2 deficiency enhances the sensitivity of cells to EZH2 inhibitors, which suggests the potential of this combination therapy[231].

Moreover, targeting SphK/S1P signaling represents an innovative strategy for anticancer therapy that has shown efficacy in in vitro studies and in tumor xenograft models. First, the inhibitor of SphK ABC294640 inhibits CRC cell growth in vitro and in vivo[210] and colitis-driven colon cancer in mice[211]. ABC294640 is currently under investigation in phase Ib and phase II clinical trials for the treatment of patients with solid tumors[232]. Notably, FTY720, a sphingosine analog derived from myriocin, curtails the amplification loop involving SphK1, S1P, and S1PR1, which results in the elimination of the NF-kB/IL-6/STAT3 amplification cascade and colon cancer development[233]. In addition, treatment of resistant CRC cells with fingolimod (FTY720), an S1PR antagonist, results in resensitization to cetuximab both in vitro and in vivo, which indicates that the combination of cetuximab and fingolimod is an effective therapeutic strategy in cetuximab-resistant CRC[234]. The metabolic rewiring of phospholipid pathways in CRC represents a hallmark of tumor plasticity and therapeutic resistance. Key enzymes such as LPCAT3, LPCAT2, and ChoKα drive lipid remodeling, chemoresistance, and stemness by modulating phospholipids. Pharmacological targeting of these nodes not only suppresses tumor growth but also restores chemosensitivity, as exemplified by the synergy between MN58b and 5-fluorouracil. The anticancer drugs that target phospholipid metabolism are detailed in Table 2.

Targeting cholesterol metabolism in CRC

Rapidly proliferating CRC cells exhibit increased cholesterol dependency to fuel membrane biogenesis and oncogenic signaling via cholesterol-derived metabolites (bile acids and steroid hormones)[235]. Consequently, the depletion or blockade of cholesterol trafficking hinders tumor growth and invasion in CRC[43].

Due to the vital functions of cholesterol in cancer progression, impeding active cholesterol synthesis may be an effective CRC treatment. Statins serve as first-line pharmacological agents that target HMGCR to inhibit cholesterol synthesis, and emerging evidence supports their pleiotropic antitumor effects. A large population-based study demonstrated that statin use is protective against CRC[236]. In another study, patients who received statins for more than 5 years even had a 47% lower risk of CRC than nonstatin users[213]. Notably, lovastatin exerts dual inhibitory effects on oncogenic signaling pathways and suppresses CRC progression through coordinated blockade of the Wnt/β-catenin and Hippo/yes-associated protein-TAZ axes[237]. Additionally, other enzymes involved in cholesterol biosynthesis can also be targeted in CRC treatment. Ro 48-8071, an inhibitor of oxido squalene cyclase, has shown significant efficacy in suppressing the growth and metastasis of CRC[214]. Moreover, the antitumor effects of Ro 48-8071 are increased when combined with 5-fluorouracil[214].

As discussed above, cholesterol esterification plays a role in CRC progression. Aberrant accumulation of cholesteryl ester is mediated by ACAT1. Avasimin, a potent inhibitor of ACAT1, has been shown to suppress the proliferation of CRC cells in vitro and in xenograft models[215]. Moreover, ACAT1 inhibition synergizes with 5-fluorouracil to enhance the antiproliferative effects of this drug, which suggests the therapeutic potential of this combination[238]. Thus, the inhibition of ACAT1 can play an antitumor role and enhance chemotherapy efficacy.

In addition to blocking the biosynthesis of cholesterol and its derivatives, cholesterol efflux has been investigated in CRC. LXRs are nuclear receptors that play important roles in maintaining the metabolic balance of cholesterol. Its agonists have shown promising results for the treatment of a variety of cancers. In preclinical studies, treatment with the specific LXR agonist GW3965 suppressed the proliferation of colon cancer cells. GW3965 Leads to the strong induction of known LXR target genes, including ABCA1/ABCG1, SREBP1c and SCD1[239]. Pharmacological inhibition of ABCA1, which mediates cholesterol efflux, has also shown promise in preclinical models. The natural compound iso liquiritigenin directly inhibits ABCA1, which effectively reduces PGC-1α-driven CRC metastasis[240]. Similarly, silencing ABCA1 via RNA interference suppresses the proliferation of LDL-1 cells, which supports its role as a viable therapeutic target[241]. The anticancer drugs that target cholesterol metabolism are detailed in Table 2.

LD accumulation and CRC

The abnormal accumulation of LDs occurs in various tumors and is increasingly recognized as a hallmark of cancer. Dysregulation of LDs and their related enzymes is implicated in CRC. Mechanistically, LDs serve as a critical reservoir of FAs that cancer cells can use, particularly in cases of increased demand for lipids.

The key proteins, DGATs, which are involved in lipid accumulation, the promotion of LD formation and protection against lipid peroxidation in cancer, have been found to play indispensable oncogenic roles. Specifically, obesity exacerbates CRC progression by upregulating diacylglycerol O-acyltransferases (DGAT1/2) via the FOXO3/MYC axis[242]. Moreover, DGAT1/2 inhibition improves FOXO3 activity by attenuating PI3K, which results in reduced MYC-dependent DGAT2 expression and less accumulation of LDs. Further analysis revealed that this inhibition attenuates the growth of colon cancer cells and colonospheres via FOXO3/p27kip1 cell cycle arrest and reduces the number of colonic tumors in APC^Min/+ mice fed a HFD[242]. adipose triglyceride lipase, the key enzyme needed for LD mobilization, also plays a crucial role in the development of obesity-driven colon cancer[243]. In addition, PPAR-α is activated in CSCs, in which LDs accumulate; however, LDs do not accumulate in non-CSCs, and pharmacological and genetic inhibition of PPAR-α suppresses cancer stemness[244]. In conclusion, these findings suggest that LD synthesis and utilization in CRC are promising as a basis for effective treatments[245].

LDs also modulate chemoresistance. Prothymosin-α induces LD accumulation by activating SREBP-1-dependent lipogenesis and STAT3, which confers resistance to gemcitabine[246]. Furthermore, LPCAT2-mediated LD accumulation impairs caspase cascade activation and ER stress responses. Notably, this droplet accumulation is associated with a reduction in immunogenic cell death and CD8 + T-cell infiltration in mouse tumor grafts and metastatic tumors of CRC patients, thus supporting CRC chemoresistance[222]. Emerging evidence highlights a distinctive oncogenic mechanism of REG4 in colorectal carcinogenesis characterized by LD-mediated chemoresistance[247]. Notably, REG4 orchestrates metabolic reprogramming in CRC cells by driving LD biogenesis, a critical adaptive strategy to circumvent chemotherapy-induced cytotoxicity. At the molecular level, REG4 executes dual regulatory mechanisms to suppress de novo lipogenesis: (1) Transcriptional silencing through epigenetic modulation by disrupting the interaction between SREBP1 and histone acetylation complexes at the promoter regions of ACC1 and ACLY; and (2) Posttranslational regulation by promoting the ubiquitin-mediated proteasomal degradation of the ACC1 and ACLY enzymes. This dual-tiered regulatory network underscores the pivotal role of REG4 in lipid metabolism reprogramming and chemoresistance development in CRC.

These findings collectively position LD synthesis and utilization as actionable therapeutic targets. Strategies such as DGAT1/2 inhibition, PPAR-α blockade, or disruption of LD-STAT3/SREBP-1 crosstalk may reverse chemoresistance[242,244,246]. However, challenges remain in addressing context-dependent LD heterogeneity and optimizing combinatorial regimens to overcome compensatory metabolic pathways. Future studies should prioritize the translation of LD-targeting agents into clinical trials to leverage biomarkers such as DGAT1/2 or PPAR-α to stratify patients and refine therapeutic efficacy. By dismantling the LD-driven metabolic armor of CRC cells, these approaches hold promise for overcoming one of the most recalcitrant features of this malignancy.

Lipid metabolism and therapy resistance

As previously demonstrated, targeting lipid metabolism has emerged as a novel therapeutic strategy for CRC. However, drug resistance has progressively become a key factor that limits treatment efficacy. Mechanically, lipid metabolism constitutes a complex and highly redundant network, wherein the inhibition of a single key enzyme may activate compensatory alternative pathways. For example, while pharmacological inhibition of SCD represents a promising therapeutic strategy, current inhibitors have shown limited clinical efficacy in cancer treatment[53]. This suboptimal response suggests that SCD inhibitor-resistant tumors may activate compensatory FA desaturation pathways to maintain the biosynthesis of critical lipid species. Notably, FADS2 has been identified as a key mediator of FA desaturation in SCD inhibitor-resistant cancers and in primary tumors[54]. Similarly, enhanced cholesterol biosynthesis and increased cholesterol uptake can occur concomitantly in cancer cells. In this context, the suppression of HMGCR in the cholesterol biosynthesis pathway may trigger compensatory mechanisms, including increased cholesterol uptake or de novo synthesis of alternative lipid species, to preserve membrane integrity and cellular function. As cancer progresses, the TME also undergoes lipid metabolic reprogramming. CAFs, an important stromal cell type in the TME, are activated by TGF-β and LPA signaling in the TME. Recent metabolomics evidence indicates that lipids transferred from CAFs increase tumor cell lipid uptake, thereby alleviating the metabolic stress induced by lipid metabolism inhibition[176].

On the basis of the comprehensive understanding of lipid metabolism, combined inhibition of multiple lipid metabolism-related targets may overcome the drug resistance associated with single-target inhibition. Moreover, emerging evidence has demonstrated that combination approaches that integrate lipid metabolism modulation with conventional tumor chemotherapy or targeted therapies significantly enhance antitumor efficacy[248]. Notably, the dual-targeting of FA transporters[249] or rate-limiting enzymes[197] in conjunction with immune checkpoint blockade has shown synergistic increases in antitumor immune responses. These findings establish a rational framework for optimizing therapeutic strategies in oncology.

CONCLUSION

Lipids play pivotal roles in CRC progression, as they function not only as energy reservoirs and structural components of membranes but also as critical signaling molecules that drive oncogenic pathways. Tumor cells undergo extensive lipid metabolic reprogramming to adapt to the nutrient-deprived and hypoxic TME; this reprogramming is characterized by increased de novo lipogenesis, FAO, and exogenous lipid uptake through transporters such as CD36 and FABPs. Furthermore, lipid metabolic rewiring in CRC cells profoundly reshapes the TME, fostering immunosuppression through lipid-laden immune cell dysfunction. For example, CD8 + T cells exhibit exhaustion due to CD36-mediated oxidized lipid uptake and ROS accumulation, whereas Tregs and TAMs exploit FAO and cholesterol synthesis to maintain immunosuppressive phenotypes[11,13,166]. CAFs contribute to CRC migration by supplying lipids to tumor cells[155]. Targeting lipid metabolism is a promising therapeutic strategy, as evidenced by the preclinical successes of ACLY inhibitors (ETC-1002), FASN blockers (TVB-3664), and CD36-neutralizing antibodies, which synergize with chemotherapy to suppress metastasis and stemness[178,188,193]. Statins and ACAT1 inhibitors disrupt cholesterol homeostasis, whereas DGAT inhibitors impair LD formation, which shows the potential of combination strategies[215,236,242]. However, clinical translation remains challenging due to metabolic redundancy and compensatory pathways (such as FADS2-mediated desaturation, which bypasses SCD1 inhibition)[53,54]. Chemoresistance occurs when chemotherapy is still the first-line therapy for cancer treatment. LDs play an important role in tumor chemotherapy resistance, as disrupting LD accumulation reverses chemoresistance[242], which suggests the great potential of targeting LDs to ameliorate chemotherapy resistance. In conclusion, lipid metabolic reprogramming is a hallmark of CRC and influences tumor initiation, immune evasion, and therapeutic resistance. Targeting key nodes involved in lipid synthesis, oxidation, and transport, particularly in combination with immunotherapy, offers a transformative approach to CRC treatment. While preclinical models highlight the efficacy of metabolic-immune combination regimens, future studies should prioritize the clinical translation of these strategies while addressing metabolic heterogeneity and microenvironmental crosstalk.