Solid stress is the mechanical force (compression, stretching, and shearing) contained in the ECM and cells and is transmitted through solid and elastic elements. The solid stress increases with increasing tumor size. The increase in tissue volume is a result of cell infiltration, cell proliferation, and matrix deposition. This augmented volume exerts pressure, consequently generating solid stress in the tumor and surrounding tissues. Helmlinger et al. first proposed the effect of solid stress on cancer cell biology and reported that accumulated solid stress inhibited the growth of tumor spheres (40). These pressures are large enough to compress or even destroy the blood and lymphatic vessels (41–43). Vascular compression leads to hypoxia (43, 44) and interferes with the efficacy of radiotherapy, chemotherapy and immunotherapy (45–47). Solid stress may also have additional direct effects on tumor biology, such as promoting the aggressiveness of cancer cells (48) and stimulating tumorigenic pathways in the colon epithelium (49). Elevated solid stress can regulate fluid stress by compressing the blood and lymphatic vessels within the tumor. Vascular compression reduces tumor perfusion, whereas compression of lymphatic vessels impedes the tumor’s ability of tumors to expel excess fluid from the interstitial space, resulting in an even increase in interstitial fluid pressure (50).

Normal interstitial fluid pressure homeostasis generally involves blood entering through arteries and veins, blood arriving through veins through arteries and lobes, and excess tissue fluid being expelled through lymphatic vessels. The presence of tumor tissue disrupts this homeostasis, showing high resistance to blood flow, low resistance to transcapillary fluid flow, and impaired lymphatic discharge (45, 51), resulting in increased interstitial fluid pressure (IFP). A high IFP hinders drug penetration to tumor sites, reduces the utilization rate, increases drug resistance, and affects the efficacy of radiotherapy, chemotherapy and immunotherapy (52).

In the human body, from every organ to every cell, there is a specific arrangement of microstructures, and constant normal evolution optimizes the stability, efficiency and function of tissues. Pathological changes disrupt these microstructures, leading to disturbances in homeostasis and facilitating the onset of various diseases. For example, the occurrence of atherosclerosis prevents the normal flow of blood, possibly leading to a lack of normal fluid shear force, eventually resulting in changes in the morphology and function of endothelial cells, and promoting vascular proliferation (53). Throughout the epithelial-mesenchymal transition process, epithelial cells undergo a loss of cell polarity. This results in a shift from the epithelial phenotype, which is anchored to the basement membrane, to a mesenchymal phenotype characterized by enhanced invasion and migration capabilities (54). In addition, excessive tumor growth eventually leads to abnormal collagen cross-linking, persistent stiffening of the stroma, and alterations in the TME, and ultimately facilitating immune evasion and resistance to therapy (55).

A stiff ECM activates signaling pathways, such as the FAK, MAPK, and PI3K/Akt pathways in tumor cells, and enhances their proliferation and survival capabilities (56, 57).

Increased matrix stiffness can promote the migration and invasion of tumor cells, making it easier for them to penetrate the ECM and enter blood or lymph vessels, leading to tumor metastasis (58, 59). Stiffness-mediated downregulation of the antiangiogenic isoform of VEGF, which results in the alternative splicing of more proangiogenic isoforms (60), could play an important role in regulating angiogenesis. The presence of laminin β1 chains in the ECM increases cell–cell contact during tube formation (61). In contrast, collagen I ensures the disruption of cell–cell connections (62). Notably, since collagen I fis the main component of many surrounding tissues, it promotes the migration behavior of the cells.

Monocytes and their derived macrophages, which are involved in tissue repair and remodeling, are responsive to various mechanical microenvironments. The expression of proinflammatory genes and cytokines, such as NOS2, IL-12β, IL-6, and IL-8, is upregulated in human monocytes encapsulated in agarose and exposed to a combination of shear and compression conditions. Monocyte activation tends to be associated with more M1-like phenotypes, highlighting the response of human monocytes to mechanical stimuli (65).

Hypertension and endothelial mechanical stretching have been reported to regulate the phenotype and function of monocytes. Coculture of human monocytes with fused human aortic endothelial cells under cyclic stretching of 5% or 10%, similar to hypertension, affected the distribution of human circulating monocytes, with the percentage of classical monocytes decreasing and the proportion of intermediate and nonclassical monocytes increasing. Additionally, the expression of IL-6, IL-1β, IL-23, CCL4 and TNF-α were increased in monocytes, leading to the promotion of monocyte differentiation and activation (66).

Macrophages are among of the most predominant immune cells within the TME and promote tumor growth and immune suppression (67, 68). They facilitate tumor growth by promoting neoangiogenesis, remodeling the matrix, and inhibiting tumor immunity and other mechanisms (67). While they promote ECM remodeling, macrophages are affected by changes in substrate stiffness. In addition, macrophages exhibit mechanical sensitivity to the hardening mechanical microenvironment and can respond to variations in matrix stiffness by altering the area, phenotype, migration rate and mode, function and actin tissue regulation. On rigid (150 kPa) substrates, the stiffness and phagocytosis capacity of macrophages increase, and further studies have indicated that the function of macrophages is regulated by actin polymerization (68).

Studies have shown that substrate stiffness strongly influences the phagocytic function and polarization phenotype of macrophages, and the specific influence is mainly determined by the origin of the macrophages, the biomaterial model used and the different chemical stimuli used (63, 68–71). Rukmani Sridharan et al. showed that the phagocytosis and migration of macrophages were impaired by stiff gel, and the migration mode was mainly mesenchymal, which was different from that of a softer matrix, and involved mainly ROCK-dependent deformation (63). Cougoule et al. suggested that there were differences in the migration patterns of macrophages in different substrate forms, such as amoeba migration in a porous matrix (e. g., fibrillary collagen type I) and mesenchymal migration in a dense matrix (e.g., matrix gel) accompanied by matrix proteolysis (72). Moreover, the migration rate of human monocyte-derived macrophages on flat fibronectin-coated PA gel was positively correlated with stiffness (73).

T cells are mechanically sensitive throughout their life cycle and are always exposed to different mechanical microenvironments, ranging from soft tissues such as the thymus or bone marrow to rigid tissues such as inflammatory and tumor tissues, which affects the function of T cells.

TCRs act as essential mechanical sensors in T cells, and mechanical forces initiate TCR signaling by acting directly on the TCR complex rather than on other surface receptors (74–76). ECM stiffness can greatly increase the activation of T cells (29) by extending the lifetime of the ligand-T cell receptor (TCR) bond, and triggering the influx of calcium ions (77, 78).

The migration, cytokine secretion, metabolism and cell cycle progression of human CD4⁺ T cells vary with substrate stiffness (76, 79). Migration was monitored on PA-hydrogels of various stiffness coated with biotinylated ICAM-1. CD3 and CD28 antibodies were used to activate T cells. The mean instantaneous velocity and migration distance of T cells on 100 kPa PA gels were significantly greater than those on 0. 5 kPa and 6. 4 kPa gels. With increasing rigidity, TCR/CD3, the main rigid sensing receptor, induced stronger signal transmission and gene expression. The expression of cytokines, T-cell surface markers, T-cell-specific transcription factors (TBX21 and Foxp3), and the proliferative transcription factor MYC increased with increasing stiffness in the presence of the CD3. However, it is interesting to note that some T-cell functions such as cytokine signaling, and T-cell activation, can be triggered at a lower stiffness value range (0.5– 6.4 kPa), whereas others (respiratory electron transport and glycolysis) require greater stiffness (6.4 –100 kPa). Moreover, the TCR activation induced glycolytic metabolism, the cell cycle, and the proliferation of T cells increase in response to stiff substrates (79).

The induction of human Treg cells is mechanosensitive and dependent on oxidative phosphorylation (OXPHOS). Specifically, Treg induction and metabolism are enhanced on stiffer substrates (80). In aged skin, a more aligned ECM resulting from the loss of the hyaluronic and proteoglycan link protein HAPLN1 impedes CD8⁺ T cell migration while promoting Treg infiltration in melanoma (81). Additionally, activation of integrin α4β1 has been shown to enhance the immunosuppressive capacity of Treg cells, whereas the loss of talin—an integrin-binding protein—can lead to severe systemic autoimmunity (82). Notably, collagen, a primary component of the ECM, can increase the expression of Treg biomarkers such as CD4, FOXP3, and CD25, thereby supporting the immunosuppressive TME (83).

Furthermore, the inducible co-stimulatory molecule (ICOS), a member of the CD28/CTLA4 family, is expressed on activated T cells (84, 85). ICOS-mediated costimulation facilitates the production of cytokines such as IL-4 and IL-10, suggesting the role of ICOS in supporting secondary, memory, and effector T cell responses (86, 87).

B cells sense antigens through B-cell receptors (BCRs) and react differently, contributed by the varying rigidity of antigens presented on substrates (88, 89). For example, virus particles exhibit greater stiffness (45–1,000 MPa) (90), most mammalian cells have medium stiffness (0.01–1,000 kPa) (91), and the secreted soluble pathogen antigens have a low stiffness of less than 100 Pa (92). Substrate stiffness guides B cell activation, proliferation, class switching and the antibody response in vivo (88, 93).

Dendritic cells (DCs), specialized immune cells, scan for foreign bodies or abnormal cells in the surrounding tissue. Once they recognize this danger signal, they travel to the lymph nodes to activate T cells and then trigger an immune response. To do this, dendritic cells travel long distances in the body and encounter a variety of microenvironments with different mechanical properties, such as tissue stiffness (94).

Studies have shown that mechanical stiffness is a key physical cue affecting DC differentiation/maturation, phenotype, metabolism, quality and function (94, 95). The expression of C-type lectins on immature DCs (IDCs) is regulated by substrate rigidity, leading to the internalization of different antigens. In addition, substrate rigidity impacts β-2 integrin expression and foot formation in IDCs, thus affecting cell functions (94).

DCs respond to increased stiffness through both functional and metabolic reprogramming. Higher stiffness upregulated glucose metabolism in DCs to support their inflammatory phenotype. Stiffness bolstersBMDC differentiation in vitro. Tension effects on DCs are potentially transient and reversible, to allow for the regulation of DC activation. In the course of immunotherapy, high-tension DC cells enhance the main adaptive immune capacity for tumor clearance. This effect does not require pattern-recognition receptor (PRR) ligation. In addition, rigid substrates can result in crosstalk of the adaptive immune system, contributing to diseases such as diabetes and pancreatitis. A major hippo signaling factor, TAZ, and a mediator of ion homeostasis, including PIEZO1, a potential tension sensor in DCs, regulate the metabolic function of DCs. Tension also influences the phenotype of human monocyte-derived DCs (95).

Preclinical experiments have shown that targeting myofibroblast apoptosis is an effective antifibrotic treatment. The monoclonal antibody C1-3 specifically targets transmembrane proteins expressed by hepatic myofibroblasts, and when combined with gliotoxin, it can induce the apoptosis of myofibroblasts and significantly reduce the severity of fibrosis (125). In addition, the abnormal transformation of fibroblasts into myofibroblasts can be inhibited. In this process, signals such as reactive oxygen, microRNAs, chemokines, and cytokines, which are important mediators of the phenotypic transformation of myofibroblasts, can be used as potential therapeutic targets (126–129) to inhibit the formation of myofibroblasts. MRG-201, a drug similar to miR-29, was recently tested in a phase 1 clinical trial for antifibrosis therapy (129) (NCT02603224). However, normal wound healing and other physiological functions require the critical involvement of myofibroblasts, and it has also been shown that depletion of myofibroblasts in the stroma leads to increased tumor aggressiveness and decreased survival (130), which is counterproductive.

TGF-β is widely involved in the occurrence and development of fibrosis in different organs. TGF-β is also a well-studied profibrotic cytokine. During the initiation of fibrosis, the overproduction of TGF-β or the enhancement of its profibrotic effect leads to an abnormal wound healing response. Moreover, TGF-β within the TME has been implicated in promoting immunosuppression, promoting angiogenesis, and epithelial-mesenchymal transition (EMT) (131–133). By targeting TGF-β, not only can the ECM be remodeled, but the TME can also be optimized to facilitate more effective cancer therapy outcomes. At present, many drugs targeting TGF-β, such as drugs prepared from peptides, antisense oligonucleotides, small molecule inhibitors, monoclonal antibodies and vaccines, have been developed and have entered phase I, II, and III clinical trials (134).

The ability of pirfenidone to inhibit TGF-β has been confirmed in clinical trials (135, 136), and it was the first drug approved for idiopathic pulmonary fibrosis (IPF) treatment in Europe and was in phase III trials in the United States (134, 137, 138) (NCT01366209, NCT00287729, NCT00287716, and NCT01504334). Belagenumatucel-L is an antisense prepared as an enhanced tumor vaccine that is actually a genetically engineered non-small cell lung cancer tumor cell line with better activity than a conventional tumor vaccine vaccination (139, 140). Significantly increased survival was found in patients treated with each dose of this vaccine, who entered a phase II/III clinical trial (141) (NCT00676507). In addition, the angiotensin receptor 1 blocker losartan reduces collagen I and HA by inhibiting TGF-β (44). In a preclinical model of PDAC, the strategies of losartan in reducing solid pressure, decompressing blood vessels, enhancing chemotherapy, and improving overall survival are currently being tested in a randomized clinical trials (142) (NCT01821729).

CTGF mediates the expression and signaling of TGF-β, which circulates through the TGF-β pathway and subsequently induces additional TGF-β production (143). CTGF, which is essential for TGF-β-mediated fibrosis, binds directly to TGF-β to enhance receptor association (144). Therefore, CTGF promotes ECM remodeling and fibrosis pathology by indirectly regulating ECM synthesis and MMP expression in myofibroblasts.

FG-3019 is a full-human monoclonal antibody against CTGF. Preclinical studies have shown that FG-3019 can penetrate tissues and reduce the effective tissue level of CTGF, thereby reducing profibrotic factors, rebalancing ECM secretion and processing, and restoring tissue homeostasis (144, 145). FG-3019 has been evaluated for the treatment of pulmonary fibrosis and has shown good safety and tolerability, as well as good results in imaging changes in pulmonary function and the degree of pulmonary fibrosis (144, 146) (NCT01262001, NCT00074698, NCT01890265, NCT04419558, and NCT03955146). In preclinical trials, FG-3019 combined with gemcitabine was found to promote tumor stability and prolong survival, with better efficacy than any single treatment (147). Therefore, FG-3019 was added to gemcitabine and erlotinib in a follow-up study in which naive patients with locally advanced or metastatic pancreatic cancer were recruited, and the results revealed good safety with significantly better overall survival (148) (NCT01181245). In addition, the research results support that FG-3019 has a good effect on the treatment of skin fibrosis, but it has not entered into clinical trials (145).

Lysyl oxidase (LOX), a typical member of five secretory copper-dependent enzyme families, promotes covalent cross-linking through the oxidative deamination of peptidyl lysine residues in collagen and elastin, thereby reshaping the stiff extracellular matrix (149, 150). LOX is now being recognized as a promising therapeutic target because of its dual involvement in the tumor stroma and premetastatic niche formation (151). β-Aminopropenitrile (β-APN) is a nonspecific, irreversible inhibitor of the lipoxygenase family that covalently binds to the active site of the lipoxygenase family of enzymes and is the first widely used LOX family inhibitor; however, its use was discontinued because of its high toxicity in clinical trials (152). GS-6624, a monoclonal antibody against LOX 2 (LOXL2), was well tolerated in the first half of the phase II safety study, but patients with cancer and fibrosis disease did not benefit from the phase II clinical trial (153–155) (NCT01362231, NCT01769196, NCT01479465, and NCT01472198).

In the race to create effective LOX inhibitors, a considerable challenge is that the complete crystal structure of mammalian LOX remains unknown (156). Therefore, other approaches have been proposed, including inhibition of LOX transcription factors or prevention of BMP-1 posttranslational cleavage of precursor peptides (156). The depletion of copper LOX catalytic sites in the nonspecific copper chelator tetraithiomolybdate reduced the serum LOXL2 concentration in patients with moderate- to high-risk primary breast cancer in a phase II clinical trial (157) (NCT00195091). Tetrathiomolybdate has also been used to treat idiopathic pulmonary fibrosis (NCT00189176), primary biliary cirrhosis (NCT00805805) and non-small cell lung cancer (NCT01837329). Overall, compared with β-APN, tetrathiomolybdate therapy is favorable because of its simple oral administration route, excellent tolerability, and stronger LOX inhibition (156, 157). PXS-5120A (158) and PXS-5153A (159), which are halinated allylamine drugs, have been shown to reduce collagen cross-linking, reduce the degree of liver and pulmonary fibrosis, and improve liver and lung function (158, 159). A phase I clinical trial of an oral LOXL2 inhibitor, PXS-5382A, was completed in healthy adults (NCT04183517). PAT-1251, a highly specific small molecule inhibitor of LOXL2 based on benzylamine and 2-substituted pyridine-4-methylamine, has not been tested in a phase II clinical trial to date, although it was found to be well tolerated and successfully passed a phase I clinical trial (NCT02852551). A preclinical study of the aminomethiophene-based LOX inhibitor CCT365623 demonstrated that inhibition of LOX led to delayed tumor development and reduced lung metastasis in a mouse model of breast cancer (160–162). Furthermore, CCT365623 has been shown to have good stability and specificity for LOX (160–162) but has not yet been tested in a clinical setting.

MMPs, which are secreted by tumor cells, stromal cells and other cells, play crucial roles in selectively cleaving ECM components. Their capacity to cleave and activate growth factors, chemokines, cytokines, and receptors is closely linked to the metastasis cascade and tumor angiogenesis processes, ultimately driving cancer progression (163, 164). MMPs can activate TGF-β, CTGF, KGF, macrophage inflammatory protein (MIP), bone morphogenetic protein (BMP), and other factors that are crucial for tumor progression and immune regulation. Additionally, MMPs can cleave proteoglycans, such as syndecans and glypicans, which are important for cell adhesion, signaling, and ECM organization. The release of these factors can further modulate the immune response and influence the infiltration and function of immune cells within the tumor microenvironment (165–167).

Efforts to target MMPs to combat cancer metastasis have been extensively pursued in clinical trials but have ultimately been proved unsuccessful in patients. Hence, strategies that involve modifying MMP activity to reduce ECM stiffness should be approached with caution (168, 169).

Integrins, as the connecting proteins, play direct connections from the matrix to the intracellular space (Table 2). In general, integrins serve as bridges for cell-ECM interactions, and their activity is influenced by ECM stiffness. As initiators of the mechanical sensing signaling pathway, activated integrins are able to activate myosin to generate force and then transduce mechanical signals to the nucleus to activate YAP/TAZ. Moreover, it can activate the mechanical sensor FAK and initiate downstream signaling cascades such as Src and Rho/ROCK, acting as a feedback loop to promote stromal stiffening and further worsen tumors (170). Preclinical studies have also shown that integrins can promote the progression of malignant tumors, including invasion, metastasis and drug resistance. Therefore, the specific targeting of integrins to directly block mechanical sensing in and out of cells is an attractive target.

Immune cells utilize integrins to facilitate interactions with cell adhesion molecules, which is essential for communication with other cells and the ECM ( Figure 2 ). Notably, the integrin α4β1 acts as the primary receptor for VCAM-1 on leukocytes (171). Moreover, α4β1 and αxβ2 can collaborate to bind VCAM-1, significantly enhancing leukocyte adhesion (172). Additionally, macrophage integrins α4 and α9 are pivotal in promoting both macrophage migration and survival (173). On the other hand, the activation of integrin αVβ3 in macrophages can sustain chronic inflammatory responses in pathological conditions. Conversely, the loss of αVβ3 ligation allows macrophages to exit the inflammatory state, highlighting its role in inflammation modulation (174).

Integrins also play a vital role in T cell functionality. For instance, blocking αvβ6 can inhibit SOX4 expression and enhance T cell-mediated cytotoxicity in response to immune checkpoint inhibitors, particularly in triple-negative breast cancer mouse models (175). Furthermore, integrin αvβ8 is predominantly expressed in CD4⁺CD25⁺ T cells within tumors. The specific deletion of β8 from T cells can mitigate TGFβ-mediated inhibition of CD8⁺ T cells, thereby restoring their tumor-killing capacity and synergizing with immunotherapies (176).

Monoclonal antibodies, such as LM609, are among the first integrin antagonists to be developed and have been reported to have antiangiogenic effects in preclinical models (177). Its humanized version, etaracizumab (MEDI-522), is one of the first integrin antagonists to enter clinical trials because of its efficacy in preclinical studies and has completed a phase II clinical trial in malignant metastatic melanoma (178) (NCT00066196). The human Avintegrin-specific monoclonal antibody CNTO95 against the αVβ3 and αVβ5 integrins exhibited good safety in phase I and II clinical trials and demonstrated antitumor activity (179–181) (NCT00888043, NCT00246012). CNTO95 and etaracizumab are being evaluated in further clinical trials. Cilengitide, an inhibitor of the αVβ3 and αVβ5 integrins, has completed phase II trials in patients with recurrent prostate cancer (182) (NCT00103337, NCT00121238) and NSCLC (183) (NCT01118676, NCT00842712) and is currently undergoing phase II and III trials in glioblastoma (184, 185) (NCT00813943, NCT00093964, NCT00689221). In nonclinical studies, αVβ6 integrins have been shown to inhibit the activation of TGF-β in nonclinical studies (186). GSK3008348, a small molecule inhibitor of αVβ6 integrin and the first inhaled compound of this class of drugs, is safe and well tolerated by inhalation administration. A phase I clinical trial of GSK3008348 for the treatment of idiopathic pulmonary fibrosis has been completed (187) (NCT03069989, NCT02612051).

FAK, a cytosolic nonreceptor tyrosine kinase, is activated by integrin clustering and functions as a key regulator of focal adhesion dynamics ( Figure 3 , Table 2) (188). It plays a critical role in cellular responses to ECM stiffness, making it a promising target for inhibiting such mechanotransduction pathways (189). The phosphorylation of FAK is increased in response to increased matrix stiffness, with constitutive phosphorylation observed in myofibroblasts (188, 190). In addition, our previous data revealed that FAK inhibition alters the fibrotic and immunosuppressive TME in pancreatic cancer and renders tumors responsive to immunotherapy (191). As such, FAK is also a potential target to mediate matrix stiffness responses.

The first FAK inhibitor, PF-562271, was tested in a phase I clinical trial with tolerable results and controllable safety. A total of 99 patients with advanced malignant tumors were treated with PF-562271. After treatment, two-thirds of the patients were stable (approximately 6 weeks after the end of cycle 2). This first clinical trial revealed FAK as a promising therapeutic target (192) (NCT00666926). The FAK inhibitor VS6063, acquired by Verastem, has good pharmacodynamic characteristics (192) and has completed a phase I clinical trial in combination with paclitaxel in patients with advanced ovarian cancer (193) (NCT01778803). The inhibitors VS-4718 and VS-5095 also effectively target FAK kinase activity. Furthermore, the VS-4718 inhibitor is currently in clinical trials for patients with metastatic nonhematological malignancies (NCT01849744). The recently developed FAK inhibitor GSK2256098 has also been tested in clinical trials (194, 195) (NCT01138033, NCT02428270) and has completed phase II clinical trials in pancreatic cancer (194) (NCT02428270).

The FAK inhibitors examined in clinical trials to date have controlled toxicity and good safety and have shown extended progression-free survival as monotherapy inhibitors without clinical or radiological effects. Trials are underway to increase the efficacy of treatment by combining FAK inhibitors with cytotoxic chemotherapy, targeted therapy or immunotherapy.