CSCs, characterized by self-renewal capacity and therapeutic resistance, are considered major contributors to tumor recurrence and metastasis (19). In addition, CSCs highly express immune checkpoint molecules such as programmed death-ligand 1 (PD-L1) and actively secrete immunosuppressive factors, which directly inhibit T-cell function or mediate immune tolerance. Thus, CSCs participate in TIME shaping and act as a core driver of tumor immune evasion (20–23).

It has been demonstrated that CTGF binds to its receptor integrin αvβ3 to activate the downstream focal adhesion kinase-SRC proto-oncogene, non-receptor tyrosine kinase (SRC)-nuclear factor-κB (NF-κB) and extracellular signal-regulated kinase/mitogen-activated protein kinase (ERK/MAPK) signaling pathways (14,20,21). CTGF may contribute to the upregulation of stemness-associated transcription factors, including Nanog homeobox, SRY-box transcription factor 2, and POU class 5 homeobox 1 (22), which may maintain CSC-like properties and enhance tumor invasive potential under specific microenvironmental conditions.

This mechanism is supported by functional in vitro pathway activation research (24). However, the extent to which CTGF directly maintains CSC stemness in vivo remains incompletely validated. Further clarification is needed to determine whether CTGF regulates CSC properties directly or through secondary changes in the TIME.

CTGF has been implicated in the regulation of EMT (25–27). However, most evidence is context-dependent and derived from pathway activation studies, rather than direct lineage-tracing or EMT-reversal experiments (28,29). Rather than acting as a uniformly pro-EMT factor, CTGF may modulate EMT through both tumor cell-intrinsic signaling and stromal paracrine mechanisms.

Indirect evidence suggests that CTGF may promote EMT through phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) signaling. CTGF-mediated PI3K/AKT activation has been validated in retinal pigment epithelial cells (30). Although PI3K/AKT-dependent EMT has been widely reported, CTGF-specific causal induction of EMT across tumor types remains largely inferential (31–34). In addition, syndecan-1 (SDC1)+ CAF-derived CTGF has been shown to activate fibroblast growth factor receptor 3 (FGFR3) signaling in tumor cells, thereby driving EMT progression (26). Following EMT, tumor cells have been reported to exhibit reduced antigen-presentation capacity and diminished recognition or adhesion by effector immune cells. These changes may enhance phenotypic plasticity and immune evasion potential, while altered secretory profiles may contribute to TIME remodeling (35).

Notably, emerging evidence indicates that CTGF may exert context-dependent or even opposing roles in EMT regulation. In epithelial ovarian cancer cells, CTGF expression was enriched in epithelial-like cells but was reduced or undetectable in mesenchymal-like cells. CTGF knockout induced early EMT-associated changes, including Snail upregulation, cytoskeletal reorganization, ECM remodeling, increased cell stiffness, anoikis resistance, and enhanced invasiveness, while Twist1 remained inactive, suggesting a partial or intermediate epithelial-mesenchymal state rather than complete EMT transition (28). Consistently, recombinant CTGF partially reversed several CTGF-loss-associated phenotypes, supporting the notion that CTGF is required, at least in part, for maintaining epithelial architecture and restraining early mesenchymal transformation in this context (28). Therefore, CTGF may function as either a pro-EMT mediator or an epithelial phenotype maintainer depending on tumor type, cellular state, stromal context, and upstream signaling background. These divergent findings indicate that CTGF-related EMT regulation remains incompletely defined across malignancies. Such bidirectional and context-specific regulatory behavior may even lead to opposing phenotypes within the same tumor entity. To better illustrate the heterogeneity of CTGF functions across tumor types, key tumor-specific mechanisms covered in the present review are summarized in Table I.

CTGF contributes to oxidative stress and hypoxia-inducible factor-1α (HIF-1α) expression, thereby supporting a shift toward glycolysis-dominant (Warburg-like) metabolic reprogramming (18,36). Lactate accumulation resulting from this process contributes to an acidic TME and impaired immune-cell function (33)

Notably, CTGF may also suppress tumor metabolism in specific contexts. In oral squamous cell carcinoma (OSCC), CTGF was shown to decrease the extracellular acidification rate, oxygen consumption rate, ATP production and mitochondrial DNA copy number, and was demonstrated to promote ubiquitin-proteasome-mediated degradation of mitochondrial transcription factor A (mtTFA). Restoration of mtTFA rescued CTGF-suppressed glycolysis, oxidative phosphorylation, migration, and invasion. These findings suggest that CTGF may inhibit OSCC progression by disrupting mtTFA-dependent metabolic activity rather than promoting Warburg-like reprogramming (37).

Therefore, CTGF may exert context-dependent bidirectional effects on tumor metabolism. Its role depends not only on tumor type and microenvironmental context but also on whether CTGF preferentially activates HIF-1α-driven glycolytic programs or suppresses mitochondrial biogenesis via mtTFA degradation. The metabolic consequences of CTGF signaling, therefore, require tumor-type-specific validation. This bidirectional metabolic regulation further highlights the context-dependent activity of CTGF. Representative tumor-specific regulatory mechanisms covered in the present review are summarized in Table I.

In the process of shaping the TIME, CTGF may contribute to tumor cell survival and stress resistance under adverse microenvironmental conditions. It contributes to the tumor cell survival by upregulating Bcl-xL and cellular inhibitor of apoptosis protein 1 (38), and by activating pro-survival signaling pathways such as Notch1 and Snail1. These pathways collectively enhance tumor-cell fitness under stress conditions, including hypoxia and nutrient deprivation (39,40). This CTGF-associated survival advantage may enable subsets of tumor cells to persist under hostile microenvironment conditions. Consequently, it may help sustain the immunosuppressive state of the TIME and contribute to tumor progression.

ECM stiffening is predominantly driven by the excessive deposition and aberrant cross-linking of fibrous proteins such as collagen. CTGF promotes collagen deposition or cross-linking, thereby remodeling ECM architecture and increasing tissue stiffness, which provides structural support for tumor progression (43); furthermore, elevated CTGF expression is associated with, and in selected models contributes to, matrix metalloproteinase (MMP) induction. This process enables localized ECM degradation and may generate permissive tracks for tumor cell migration and invasion (44–46).

The interplay between collagen accumulation and matrix degradation creates a heterogeneous ECM landscape that facilitates tumor invasion. Increased ECM stiffness can also function as a physical barrier that restricts effector immune-cell infiltration, particularly CD8+ T cells, into the tumor core. This represents a mechanobiological mode of immune suppression that is partially independent of canonical cytokine signaling (43,47). However, whether CTGF-driven ECM stiffening directly impairs immune-cell penetration in specific tumor contexts remains to be fully validated.

CAFs are key effector cells that drive ECM remodeling and shape the TIME. CTGF contributes to CAF activation and ECM remodeling, thereby promoting the formation of dense matrix networks and reinforcing physical barriers within the TIME (9). This conclusion is supported by functional knockdown experiments in selected tumor models and correlative clinical observations (8,26). Current evidence supports this effect based on a combination of functional knockdown studies and correlative analyses (8,26,48). However, CTGF alone is unlikely to be sufficient to define lineage commitment or full functional reprogramming.

Activated CAFs can further secrete immunosuppressive mediators that impair CD4+ and CD8+ T-cell function, thereby attenuating antitumor immune responses at the cellular level (49). In gastric cancer, tumor cell-derived TGF-β1 was found to induce CTGF expression in CAFs through SRC-dependent ERK/Smad/JNK signaling pathways. CTGF targeting has been shown to reduce CAF-mediated tumor cell migration and invasion in this context (8). Notably, a conserved SDC1+ CAF subset associated with advanced tumor stage and poor prognosis drives tumor cell EMT, invasion, and lymphatic metastasis via Krüppel-like factor 6-regulated CTGF secretion and activation of tumor FGFR3 signaling, highlighting that CTGF and FGFR3 may serve as actionable stromal targets with anti-metastatic therapeutic implications across multiple malignancies (26).

Furthermore, an anti-CTGF/programmed cell death protein 1 (PD-1) bispecific antibody Y126S was shown to mediate suppression of CAF activation, reduction of collagen deposition, and downregulation of PD-L1 expression on CAFs by targeting CTGF in a pancreatic cancer model. This dual targeting strategy was found to enhance CD8+ T-cell-mediated antitumor immunity and to improve the efficacy of PD-1 blockade (48). However, the functional heterogeneity of CTGF across CAF subsets remains incompletely characterized.

The vascular system is a highly organized tubular network responsible for oxygen and nutrient delivery and metabolic waste removal. As a functional unit embedded within the ECM, its structure and function are directly regulated by the physicochemical properties of the ECM. Accordingly, vascular abnormalities represent a key feature and functional consequence of ECM remodeling in tumors. As a critical factor involved in vascular homeostasis, dysregulated CTGF signaling is associated with reduced pericyte coverage and abnormal basement membrane formation, as shown in vascular modeling systems and tumor-associated vascular analyses. These alterations contribute to vascular destabilization and structural disorganization (50).

CTGF-associated vascular dysfunction may impair effector T-cell adhesion and transendothelial migration, resulting in reduced immune-cell infiltration into tumor tissues (51). In addition, CTGF-associated vascular abnormalities may compromise tissue perfusion, promote hypoxia, suppress T-cell cytotoxic function, and facilitate the recruitment of immunosuppressive cell populations. Collectively, these effects further reinforce the immunosuppressive and immune-excluded state of the TIME.

TAMs are broadly classified into M1 and M2 subsets, which exert antitumor and pro-tumor activities, respectively. CTGF may contribute to pro-tumorigenic M2 polarization in solid tumors. Correlative findings from clinical and omics analysis indicate that high CTGF expression is associated with increased infiltration of M2-type TAMs (6). In addition, functional research suggests that CTGF contributes to macrophage polarization signaling via the hsa-miR-27a-3p/enhancer of zeste homolog 1 (EZH1)/lysine demethylase 3A (KDM3A) axis (7). However, direct in vivo causal validation across tumor types remains limited. In hepatocellular carcinoma (HCC), experimental evidence demonstrates that CTGF derived from mesenchymal-like tumor cells promotes M2 macrophage polarization. These M2-like TAMs further enhance tumor progression by secreting C-C motif chemokine ligand 18 (CCL18), forming a CTGF-M2 TAM-CCL18 positive feedback loop (53). Whether this regulatory axis is broadly conserved across solid tumors remains unclear.

MDSCs, as key immunosuppressive cells, inhibit the functions of T cells and natural killer (NK) cells by depleting arginine and generating reactive oxygen species (ROS), thereby mediating tumor immune evasion (54,55). As a critical downstream molecule of the TGF-β signaling pathway, CTGF may potentiate TGF-β-driven MDSC expansion and functional maintenance (14,56). However, this mechanism is mainly supported by pathway-level evidence and indirect inference rather than direct CTGF-MDSC functional or depletion studies. High CTGF expression is associated with increased secretion of MDSC-recruiting chemokines derived from CAFs and with MDSC accumulation in the TIME (57), suggesting a correlative link between CTGF-enriched stromal environments and MDSC infiltration rather than a fully established causal relationship. Current evidence is largely limited to correlative findings, and it remains unclear whether MDSCs and CTGF+ CAFs are spatially and functionally coupled within the TIME. The subtype-specific regulatory effects of CTGF on polymorphonuclear MDSCs and monocytic MDSCs, including their proliferation, apoptosis, metabolic programs, and suppressive effector molecule expression, remain largely undefined.

TANs predominantly exhibit the pro-tumor N2-like phenotype in the TME, contributing to angiogenesis, stromal remodeling, and immunosuppression (58). CTGF may contribute to the establishment of a pro-tumor TAN niche by reinforcing TGF-β-associated stromal remodeling and neutrophil-recruiting inflammatory programs. However, current evidence remains largely indirect, based on TGF-β-dependent TAN polarization models and CTGF-associated stromal signatures, rather than direct experimental validation of CTGF in TAN lineage commitment. TGF-β is a key regulator of TAN functional polarization, promoting pro-tumor N2-like pro-tumor phenotypes, whereas TGF-β blockade can reprogram TANs toward an antitumor N1-like state (59,60). As a downstream effector and functional amplifier of TGF-β signaling, CTGF contributes to CAF activation, collagen deposition, ECM remodeling, and tissue fibrosis (14). These stromal alterations may collectively facilitate the recruitment and functional polarization of N2-like TANs, which are associated with angiogenesis, matrix remodeling, chronic inflammation, immune suppression, and tumor progression (61).

In addition to stromal remodeling, CTGF may regulate TAN infiltration through chemokine-mediated neutrophil recruitment. TAN trafficking is largely controlled by neutrophil-attracting C-X-C motif chemokine receptor (CXCR)2 ligands, including C-X-C motif chemokine ligand (CXCL)1, CXCL2, CXCL5, and CXCL8/IL-8 (61). Evidence from inflammatory stromal models suggests that CTGF contributes to CXCL1 induction, indicating that CTGF-rich stromal environments may enhance neutrophil recruitment and sustain a pro-tumor TAN-enriched niche (62). Collectively, CTGF may contribute to tumor progression by a TGF-β/ECM-chemokine axis that supports TAN recruitment, N2 polarization, angiogenesis, matrix remodeling, and immunosuppression.

Research in chronic inflammatory disorders, such as kidney disease, have demonstrated that CTGF and its C-terminal module mediate human CD4+ T-cell polarization by driving the differentiation of proinflammatory Th17 polarization and local inflammatory responses (63). Although this mechanism was characterized in non-neoplastic inflammatory settings, given that the tumor microenvironment, such as in pancreatic cancer, shares key features with chronic inflammatory tissues, including persistent protease activity that drives ECM remodeling and tumor progression (64), it is plausible that the proinflammatory effects of CTGF fragments observed in inflammatory contexts may also contribute to shaping the T-cell landscape in cancer. In addition, CTGF contributes to CAF activation and ECM deposition, which may physically restrict infiltration of CD8+ T cells into tumor cores and impair effector function (9). CTGF expression was also shown to be positively correlated with regulatory T-cell infiltration in the TIME (6). These findings suggest that CTGF may promote immunosuppressive T-cell polarization while simultaneously limiting effector T-cell infiltration and activity through stromal remodeling. However, the precise molecular mechanisms underlying CTGF-induced T-cell dysfunction remain incompletely understood. Its potential role in regulating T-cell exhaustion, metabolic reprogramming, and spatial distribution within tumors remains to be elucidated.

B cells exert dual functions in the tumor immune microenvironment, exerting both antitumor effects and immunosuppressive functions through differentiation into regulatory B cells (Bregs) (65,66). Direct evidence for CTGF-mediated regulation of B cells remains limited. However, studies in hematologic malignancies have indicated that high CTGF expression is associated with poor prognosis in precursor B-cell leukemia (67,68). In solid tumors, CTGF-associated fibrotic and inflammatory microenvironments may promote the accumulation of MDSCs and other immunosuppressive cell populations, and MDSCs may contribute to Breg generation (69). This suggests that CTGF may indirectly regulate B-cell immunosuppressive programs through an MDSC-dependent axis.

NK cells represent the first-line effector population in antitumor immunity (70). CTGF may suppress NK-cell function primarily through indirect mechanisms. Structurally, CTGF-driven EMT and ECM remodeling may restrict NK-cell infiltration into the tumor parenchyma. Metabolically, CTGF-mediated hypoxia and lactic acid accumulation may further impair NK-cell effector functions (36,71).

Direct cytokines within the TIME mainly include TGF-β, VEGF, IL-6, IL-35, IL-1β, IL-12, TNF-α, and interferon (IFN)-γ. In tumors characterized by prominent fibrosis, such as breast cancer, CTGF and TGF-β are frequently co-localized and highly expressed (12). This co-activation can synergistically amplify profibrotic signaling, promoting CAF activation and ECM deposition (8,14), enhancing MDSC expansion and functional maintenance (14), and increasing oxidative stress (36,69,72). Accordingly, CTGF may function as a downstream effector or transcriptional target of TGF-β signaling in fibrotic and metabolically stressed tumor contexts (14,18,36,56). In these settings, CTGF can further amplify TGF-β-driven stromal and immunosuppressive programs by reinforcing CAF activation, ECM deposition, and MDSC accumulation (8,14,56). However, CTGF may also act as a relatively independent paracrine regulator when produced by specific stromal or tumor-cell subsets, such as SDC1+ CAFs or mesenchymal-like HCC cells (31,53). This may contribute to spatial segregation between tumor cells and immune effector cells, thereby impairing immune-cell function and promoting tumor progression (73).

In acute myeloid leukemia, CTGF was shown to mediate enhanced tumor angiogenic capacity through the CTGF-VEGFA axis. This axis promotes aberrant angiogenesis and suppresses endothelial adhesion molecule expression, thereby restricting T-cell adhesion and infiltration. Collectively, these effects further consolidate an immunosuppressive TIME and support tumor progression and metastasis (74). CTGF activates and upregulates the expression of IL-1β and IL-6 by activating the NF-κB signaling pathway, thereby triggering STAT3 activation, exacerbating local inflammation (10,75), and driving inflammation-cancer transformation (76). In addition, the miR-27a-3p/EZH1/KDM3A/CTGF axis was also shown to contribute to TNF-α downregulation in vivo (7). Reduced TNF-α may facilitate M2 macrophage polarization, weaken antitumor immunity, and ultimately promote tumor progression (7). However, most evidence is derived from pathway activation studies (7,10,74,75) rather than direct cytokine perturbation experiments that isolate CTGF as a primary upstream regulator. Therefore, the cytokine regulatory network governed by CTGF remains to be further validated in tumor-specific experimental models.

Indirect cytokines in the TIME mainly include IL-8 (CXCL8), granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-10, and IL-4. As a key chemokine, IL-8 has been shown to promote the recruitment of TANs and MDSCs to tumor sites via the CXCR1/2 axis, and was revealed to be closely associated with angiogenesis and the amplification of local inflammation (77,78). These myeloid populations produce immunosuppressive mediators, including arginase-1, ROS, and TGF-β, within the TIME. This suppresses CD8+ T cells, driving the transition of inflammatory signals to an immunosuppressive phenotype, and facilitating the establishment and maintenance of the TIME (79,80). CTGF can induce IL-8 expression by activating the JNK, ERK, and p38 MAPK signaling pathways and by enhancing IL-8 mRNA stability (11). These findings support the existence of a CTGF-IL-8-myeloid cell axis, which may form an inflammation-immunosuppression feedback loop within the TIME. This loop may further amplify the immunosuppressive state and contribute to the aberrant remodeling of the TIME. Beyond IL-8, emerging evidence suggests that CTGF may regulate IL-10, another key immunosuppressive cytokine in the TIME. In glioblastoma, the miR-27a-3p/EZH1/KDM3A/CTGF axis was found to contribute to IL-10 upregulation, which was associated with enhanced M2 macrophage polarization and tumor progression (7). The proposed CTGF-GM-CSF relationship remains inferential and requires direct mechanistic validation.