Tumor microenvironment and mechanotransduction pathways: Novel targets and new directions for cancer therapy

Abstract

In cancer research, the extracellular matrix (ECM) of the tumor microenvironment (TME) was once regarded as a mere passive structural scaffold for tumor tissue, while the biomechanical cues generated by its dynamic physical remodeling have now emerged as crucial active regulators that drive tumor initiation, progression and metastasis. By sensing and transducing these abnormal mechanical signals (e.g., matrix stiffness, fluid shear stress) through four core mechanotransduction mechanisms (mechanosensitive ion channels, adhesive complexes, nuclear mechanotransducers, and cytoskeletal dynamics), tumor cells convert them into intracellular biochemical responses. This review begins with the ECM within the TME, exploring mechanosensing and mechanotransduction in the TME, to construct a multi-dimensional research system integrating ECM abnormal remodeling, core mechanotransduction mechanisms and the multi-faceted regulatory effects of mechanical signals on tumor progression. Furthermore, it discusses the application of biomechanical cues in clinical tumor diagnosis, as well as attempts at anti-tumor therapies involving artificial intervention in ECM components to alter its physical properties, thereby providing reference ideas for biomechanical research.

Graphical abstract

1. Introduction

Over the past several decades, cancer has been recognized as one of the leading causes of death worldwide and a key factor limiting the extension of healthy lifespan.¹ Tumor cells, immune cells, and the ECM together constitute the living microenvironment of the tumor. Previous research has suggested that tumor cells themselves, as well as their crosstalk to immune cells, are the main factors influencing tumor growth, and therapeutic target for cancers, while the ECM merely provides physical attachment for tumor cells. In fact, the physical properties of the ECM, such as matrix stiffness, fluid shear stress, and solid stress are also important regulators of tumor progression. Specifically, mechanical forces are transmitted through the ECM to the cells within the TME and are perceived by these cells,which profoundly affect tumor cell adhesion, morphology, proliferation, differentiation, migration and invasion, as well as their interactions with other components in the microenvironment, thus forming a complex regulatory network that promotes malignant tumor progression. The core of mechanobiology lies in elucidating how cells perceive mechanical signals in the TME through mechanosensing and convert these signals into intracellular biochemical signals via mechanotransduction.² Therefore, mechanosensing and mechanotransduction, as two key processes in mechanobiology, constitute the cornerstone of research in this field. Mechanosensing refers to the process by which cells detect mechanical changes in the TME through specific mechanoreceptors. Regardless of the inherent differences among these biomechanical factors, cells recognize them via adhesion molecules such as integrins, thereby achieving attachment to the extracellular matrix and forming contact structures known as adhesive complexes or focal adhesions (FAs). Mechanotransduction is a cascade reaction process that converts perceived mechanical signals into biochemical signals, which ultimately trigger biological responses by regulating gene expression and cellular behaviors.³ This signal transmission process plays a crucial role in the development of malignant phenotypes of tumor cells, including proliferation, survival, migration, stemness maintenance, and therapeutic resistance (see Fig. 1).

Fig. 1.

ECM-derived mechanical cues (matrix stiffness, fluid shear stress, intercellular tension) are transduced via Piezo1/2-Ca²⁺, integrin-RhoA/ROCK, and nuclear mechanotransduction (Lamin A/C) pathways. These signals converge on YAP/TAZ nuclear translocation and TEAD-mediated transcription, driving malignant tumor phenotypes GAG proliferation, invasion, immune evasion, and stemness maintenance. Created with BioGDP.com.³⁵.

This review provides a systematic overview of the mechanistic cascade linking mechanical cues in the intracellular mechanotransduction signaling and the subsequent manifestation of malignant tumor phenotypes. This review constructs a comprehensive and multi-dimensional research system by integrating the abnormal remodeling of TME and ECM components, the four core mechanotransduction mechanisms of tumor cells, and the multi-faceted regulatory effects of mechanical signals on tumor progression. It also extends the research perspective from basic mechanistic exploration to clinical translation, and systematically clarifies the application value of TME biomechanics in cancer diagnosis, treatment and prognosis evaluation. In addition, it contributes to exploration of innovative therapeutic strategies based on this mechanism in depth, points out the current challenges in this field, and provides an outlook on future research directions.

2. Mechanosensing and mechanotransduction in the tumor microenvironment

2.1. Components of tumor ECM

As the hub of physical and biochemical signals in the TME, the dynamic remodeling of ECM components directly determines the generation and transmission efficiency of mechanical signals.⁴ Tumor ECM differs significantly from that of normal tissues in both composition and structure, and this difference is the fundamental cause of abnormal mechanical signals in the TME. The main components of tumor ECM include the following: collagens, proteoglycans, and fibronectins. Collagens are fibrous proteins assembled from triple-helical polypeptide chains, with type I, III and IV being the most dysregulated in tumors; their polypeptide backbones form rigid, rope-like supramolecular fibrils that constitute the ECM's structural backbone.^4,5 Proteoglycans consist of a core protein covalently linked to long, negatively charged glycosaminoglycan (GAG) chains, which are highly hydrated and form a gel-like matrix via electrostatic and hydrogen bonding.⁶ Fibronectins are dimeric glycoproteins composed of two identical polypeptide chains connected by disulfide bonds, containing multiple modular binding domains that mediate cell-matrix and matrix-matrix interactions.⁷ At the molecular level, collagens self-assemble into insoluble fibrillar networks that form the ECM's mechanical scaffold, while proteoglycans fill the interstitial spaces of collagen fibrils and fibronectins crosslink collagen fibers and bind to cell surface integrins, integrating all three components into a cohesive, interconnected ECM structure.^8,9 Collagens are the primary determinant of ECM stiffness—elevated deposition and cross-linking of type I collagen, in particular, drastically increases matrix rigidity by reinforcing the fibrillar network and enhancing intermolecular mechanical resistance.¹⁰ In contrast, proteoglycans enhance ECM flexibility: their hydrated GAG chains create a viscous, deformable matrix that dissipates mechanical force and enables the ECM to undergo reversible structural deformation under physical stress. Fibronectins modulate both stiffness and flexibility in a context-dependent manner, as their fiber tension and crosslinking degree can stiffen local ECM regions while their modular structure maintains structural pliability for cell-mediated matrix remodeling.^11,12 Collagen is the most abundant protein in the ECM, with the dysregulated expression of type Ⅰ, Ⅲ and Ⅳ being the most prominent, notably, collagens exert dual pro-tumor and anti-tumor effects in the TME(Table 1).¹³ In healthy ECM, collagens I, III and IV form a hierarchically ordered, moderately cross-linked fibrillar network with a balanced ratio, maintaining optimal stiffness, elasticity and structural integrity to support normal cell behavior. The spatiotemporal dysregulation of this protein often leads to a shift in the role of ECM during tumor progression.¹⁴ For instance, collagens act as a barrier in early-stage tumors and promote invasion after being remodeled in advanced-stage tumors; moderate collagen density restricts tumor cell dissemination, while excessively high density triggers immune evasion.¹³ The dual regulatory effects of the ECM give a rise to many intriguing questions. How do changes in ECM stiffness or ECM permeability influence tumor biological behavior and tumor immunity? Which of these factors plays a dominant role? Is it possible to modulate the physical properties of the ECM by artificially adjusting its component ratios, thereby influencing cancer development? These undoubtedly are the motivation for our research into the role of mechanotransduction signals in the TME (see Table 2).

Table 1.

The effects of different collagens on tumors.

Collagen Type	Function Type	Mechanism
COL I	Promoting Cancer	MMPs cleave it to produce cCOLI, which activates DDR1. Through the NF-κB-p62-NRF2 pathway, it promotes macropinocytosis and mitochondrial biogenesis in tumor cells, enhancing their metabolism and growth.15
		Tumor-derived COLI homotrimers activate the FAK/AKT signaling pathway via integrin α3β1 and DDR1, driving tumor proliferation and inhibiting immune cell infiltration.16
		In NSCLC, upregulated COLI forms a physical barrier that prevents T cell infiltration and attack.17
	Inhibiting Cancer	It forms a physical barrier, mechanically restricts tumor cell dissemination, and this mechanical restrictive effect can counteract some pro-tumorigenic mechanical signals.18
COL III	Inhibiting Cancer	The reduction of COLIII in the matrix increases the invasiveness of proliferative breast cancer.19
		Dormant tumor cells secrete COL3A1 (the product encoded by the COL III gene) to establish an ECM niche. It activates the STAT1 signaling pathway via the DDR1 receptor, inhibiting tumor cell proliferation and maintaining the dormant state.20
COL IV	Promoting Cancer	In certain tumors, the COLIV α5 chain drives glycolysis via c-Myc to promote tumor progression.21

Table 2.

Summary of mechanical stimuli, key sensors, downstream signaling pathways and functional outcomes in the tumor microenvironment.

Mechanical Stimuli	Key Sensors	Downstream Signaling Pathways	Functional Outcomes in TME
Matrix Stiffness (Elevated Matrix Stiffness)	Integrin、Piezo1、FAK	1.Integrin-FAK-RhoA/ROCK 2.Piezo1-Ca2+-PI3K/AKT 3.YAP/TAZ-TEAD	Tumor Cells: proliferation↑ migration/invasion↑ Immune Cells: immune evasion.10 Stromal Cells: ECM remodeling
Fluid Shear Stress	Piezo1、Integrin	1.Piezo1-Ca2+-AKT/ERK 2.Integrin-Gq/G11-FAK-NF-κB 3.Ca2+-NF-κB Integrin-	Tumor Cells: apoptosis↑, viability↓ Immune Cells: MHC-II↑,(CD80/CD86)↑ antigen presentation ↑ Vascular Cells: tumor angiogenesis.↓22,23
Intercellular Mechanical Tension	E-cadherin/cadherin, integrin, cytoskeleton (actin-myosin)	1.E-cadherin-RhoA/ROCK 2.E-cadherin-α-catenin-β-catenin	Tumor Cells: migration↑; infiltration↑; chromatin remodeling.
ECM Topography(oriented collagen fibers)	Integrin, DDR1 (discoidin domain receptor 1)	1.DDR1-NF-κB-p62-NRF2 2.Integrin-FAK-RhoA/ROCK 3.YAP/TAZ pathway	Tumor cells: invasion↑ metastasis↑; metabolism↑. Immune cells: infiltration↓ immune suppression. ↑ ECM remodeling: TME physical barrier.↑
ECM Viscoelasticity	Integrin、Piezo channels、YAP/TAZ	1.Piezo1-Ca2+-NF-κB 2.Integrin-FAK-JAK/STAT 3.RhoA/ROCK	tumor cell: adhesion↑ proliferation↑ Immune cells: immunosuppressive↑

2.2. Mechanotransduction in tumor cells

Mechanical forces play a crucial role in tumor development. Tumor cells mainly convert mechanical signals from the TME into intracellular biochemical responses through the following four core mechanisms, which collectively constitute the molecular basis of mechanotransduction.

2.2.1. Mechanosensitive ion channels

As rapidly responding mechanoreceptors, mechanosensitive ion channels can regulate ion flow by sensing changes in cell membrane tension, thereby initiating intracellular signaling pathways. Piezo channels are mechanosensitive ion channels. As direct mechanical sensors, they can be directly activated by membrane tension, enabling rapid mechano-electrical conversion. Meanwhile, Piezo channels also act as amplifiers of integrin-mediated mechanical tension. Their sensitivity is enhanced by integrin–cytoskeleton pulling, which amplifies extracellular mechanical signals. The two systems work synergistically to accomplish mechanotransduction.²⁴ A typical example is the Piezo protein channel.²⁵ Mechanical stimulation can activate two types of mechanosensitive ion channels (Piezo1 and Piezo2) on the cell membrane, induce calcium ion (Ca²⁺) influx, and mediate different downstream signaling pathways, thereby jointly promoting tumor proliferation. The specific mechanisms are as follows: In the Piezo2 pathway, channel activation and opening induce Ca²⁺ influx, which in turn activates and phosphorylates AKT. Phosphorylated AKT further phosphorylates GSK-3β, making it lose the ability to degrade Snail and leading to Snail accumulation. Snail binds to the promoter region of E-cadherin, inhibits the transcriptional expression of the epithelial marker E-cadherin, and promotes the expression of the mesenchymal marker Vimentin.²⁶ In the Piezo1 pathway, massive Ca²⁺ influx triggered by channel activation initiates multiple signals: first, it activates the PI3K/AKT/mTOR pathway and phosphorylates molecules such as AKT,²⁷ promotes Cyclin D expression and inhibits p21, thus facilitating cell cycle progression; second, it reduces cytochrome c release and inhibits Caspase-3/9 to block apoptosis; third, the decreased expression of Piezo1 regulates RNF114 via GPRL3, and assists in proliferation by affecting F-actin function. ECM stiffness-induced activation is mediated by the integrin-cytoskeleton axis: cells bind to the ECM via integrins. A high-stiffness matrix induces the coupling between cytoskeletal contractile forces and ECM counterforces, thereby elevating membrane tension and driving the conformational transition of Piezo1 from a curved “nanobowl” state to a flattened state, which in turn opens the channel. Subsequent to Ca²⁺ influx, myosin light chain kinase (MLCK) is activated, which enhances cytoskeletal contractility and establishes a positive feedback loop. Downstream, pathways including FAK and YAP/TAZ are activated to regulate cell proliferation and migration. And Fluid shear stress can act directly on the Piezo1 membrane structure to rapidly open the channel, independent of cell-matrix adhesion. Following Ca²⁺ influx, the signal is amplified through the “extracellular Ca²⁺ influx-endoplasmic reticulum Ca²⁺ release” signaling amplification cascade. Downstream, pathways such as AKT are activated to regulate endothelial cell arrangement and vasodilation. Both activation mechanisms exhibit millisecond-scale activation kinetics and are core-dependent on Ca²⁺ influx. The key distinction is their dependence on cell-matrix adhesion.

In a high matrix stiffness environment, the Piezo1 channel undergoes conformational changes, mediating Ca²⁺ influx,²⁸ activating downstream signaling pathways, and enhancing the invasive ability of tumor cells.²⁹ In glioma cells, Piezo1 regulates cell proliferation, tissue sclerosis, enhances the survival ability of circulating tumor cells, and drives the invasive progression of gliomas by sensing mechanical forces and activating downstream signaling pathways.³⁰ When Piezo1 is activated by mechanical forces, the cytoskeleton of tumor cells is strengthened, which ultimately enhances the migration and invasion abilities of tumors, further highlighting the important role of mechanosensitive ion channels in tumor progression.³¹

2.2.2. Adhesive complexes

The integrin family consists of heterodimers composed of α and β subunits and functions as a central hub for the sensing and transduction of mechanical signals (see Fig. 2). Different combinations of α and β subunits generate distinct integrin heterodimers, which allows them to accurately recognize different ligands in the ECM. Heterodimers composed of different α and β subunit combinations can accurately recognize different ligands in the ECM, such as collagens, laminins, and fibronectins, establish physical connections between cells and the matrix, and achieve initial perception of mechanical signals(Fig. 3A and B).³² Notably, the functional state of integrins directly determines whether this signal transmission can proceed: when in an inactive bent conformation, integrins fail to effectively bind ECM components or recruit cytoplasmic signaling molecules, thus blocking mechanotransduction. When the ECM stiffness in the TME increases, the functional integrin heterodimers that bind to ECM components aggregate to form FAs. Focal adhesions are large molecular complexes for mechanotransduction, which gradually recruit cytoplasmic signaling molecules such as FAK, paxillin, and talin to assemble into a dynamically structured signal transmission platform.³³ This platform is both an “sensing hub” for mechanical signals and a “molecular platform” for signal transmission downstream. Through this platform, multiple downstream pathways are activated, the RhoA/Rock pathway regulates cytoskeletal remodeling, while the YAP/TAZ and PI3K/AKT-mTOR pathways synergistically promote cell proliferation and angiogenesis. The stiffness of the ECM affects the transmission efficiency of downstream signals by regulating the size and maturity of focal adhesions.³⁴ When FAs reach a certain mature degree, activated FAK further triggers downstream signaling pathways, and further affects cytoskeletal contraction and cell migration, and further promoting the malignant behavior of tumor cells.

Fig. 2.

Mechanical stimulation triggers Ca²⁺ influx through Piezo1/Piezo2, and this process promotes cell proliferation via two distinct pathways: 1. The Piezo2 pathway: Ca²⁺ activates AKT, which inactivates GSK-3β. The inactivated GSK-3β stabilizes the transcription factor Snail, which then regulates epithelial-mesenchymal transition (EMT)-associated molecules, ultimately driving cell proliferation; 2. The Piezo1 pathway (exemplified in cancer cells): Ca²⁺ activates signaling cascades including PI3K/AKT/mTOR and MAPKs. These pathways promote cell cycle progression and suppress apoptosis, thereby facilitating cell proliferation. Created with BioGDP.com.³⁵.

Fig. 3.

A ECM binds to integrins on the cell membrane; upon mechanical stimulation, integrins transition to the activated state. Activated integrins recruit molecules including Talin, FAK, and Paxillin. This recruitment further activates RhoA/Rock, which drives cytoskeletal remodeling and ultimately promotes cell proliferation. Simultaneously, it triggers the PI3K/AKT pathway (which in turn activates mTOR) and directly regulates YAP/TAZ, collectively facilitating angiogenesis. 3B Integrins adopt a bent conformation and stay in an inactive state, thereby being unable to initiate the above-described signal transduction cascades. Created with BioGDP.com.³⁵.

2.2.3. Nuclear mechanotransducers

Nuclear mechanotransducer is the physical bridge between the nucleoskeleton and cytoskeleton, enabling the nucleus to directly perceive extracellular mechanical signals. This process leads to changes in chromatin structure and regulating gene expression and cell behavior (Fig. 4). Among them, lamin A/C and chromatin remodelers are the core mediators for this signal transmission process.³⁶ Lamin A/C is a key mediator connecting mechanical signals from the cytoskeleton to the nucleus, and can transmit mechanical signals from the cytoplasm to the nucleus.³⁷ This signal transmission process is triggered by increased ECM stiffness: ECM stiffness activates membrane-bound integrins, which then recruit cytoplasmic signaling molecules (FAK, Talin, Vinculin) to form a functional complex. This complex links to the actin cytoskeleton, and the mechanical signal is further relayed through the cytoskeleton to Lamin A/C in the nucleus. Lamin A/C then modulates the Hippo signaling pathway's activity, whose changes in turn regulate the nuclear interaction between YAP/TAZ and TEAD—this entire cascade ultimately promotes cell migration. In a high solid stress environment, lamin A/C undergoes degradation, leading to nuclear envelope rupture and damage to nuclear integrity. This change not only triggers genomic instability but also activates the DNA damage response pathway.³⁸ Therefore, we can affect the proliferation and survival of tumor cells through lamin A/C. Laminin is a heterotrimeric structure formed by α, β, and γ chains, and serves as a key bridge mediating cell-matrix adhesion.³⁹ Notably, laminins regulate tumor progression by interacting with multiple signaling pathways: the Hippo pathway is a critical one, and it also engages with the EGFR and RTK pathways. Currently, it has been found that the Linker of Nucleoskeleton and Cytoskeleton (LINC) complex can transmit mechanical signals from the plasma membrane to the nucleus. With the function of the LINC complex, the nucleus can respond to mechanical forces by altering nuclear structure, achieving chromatin reorganization, and regulating key factors such as gene location and expression.⁴⁰ Among laminins, laminin 332 (LN-332)⁴¹ and laminin 511 (LN-511)⁴² are most closely associated with tumor progression.

Fig. 4.

When the stiffness of the ECM increases, integrins on the cell membrane are activated and recruit FAK, Vinculin, and Talin to form a complex. This complex further regulates the assembly of actin filaments. Subsequent changes in the actin cytoskeleton affect Lamin A/C within the nucleus, which in turn inhibits the Hippo signaling pathway. Under this condition, YAP/TAZ is not phosphorylated; instead, it translocates into the nucleus, binds to the transcription factor TEAD to form a complex, regulates the expression of related genes, and ultimately promotes cell migration. Created with BioGDP.com.³⁵.

In intrahepatic cholangiocarcinoma (iCCA), laminin-γ2 (LAMC2, a component of LN-332) interacts with chaperones in the endoplasmic reticulum to form a “pocket” structure, which binds to unglycosylated Epidermal Growth Factor Receptor (EGFR) and promotes its translation, thereby activating the EGFR signaling pathway to drive tumor growth. In vitro and in vivo experiments have shown that iCCA cells with high LAMC2 expression are more sensitive to EGFR tyrosine kinase inhibitors (e.g., erlotinib), suggesting that LAMC2 can serve as a prognostic marker and therapeutic target.⁴³ In pancreatic adenocarcinoma (PAAD), LN-332 is significantly upregulated at both the mRNA and protein levels with a high correlation between expressions. High LN-332 expression is associated with advanced tumor grade, TP53 mutation, and receptor tyrosine kinase (RTK) pathway activation, and can predict shortened overall survival of patients.⁴⁴ In breast cancer, LN-511 can act as a ligand for integrin α6β1 and promote the self-renewal of breast cancer stem cells by activating the Hippo transducer TAZ.⁴⁵ Therefore, it can be seen that LN-511 plays a role in maintaining the stemness of cancer stem cells.

2.2.4. Cytoskeletal dynamics

Cells have the ability of self-regulation in response to the surrounding environment (e.g., matrix elasticity). The cytoskeleton plays a core role in many mechanical and biological functions, and the contractility of actin-myosin is important in cell mechanosensing, through which cells can perceive matrix stiffness.⁴⁶

This mechanosensing process links to the regulation of cytoskeletal components (microtubules, intermediate filaments, microfilaments) via Rho/GTPases: mechanical stimulation triggers the activation of Rho/GTPases, which then modulate cytoskeletal dynamics to drive cell migration. Specifically, the mechanism by which ECM activates Rho is as follows: ECM binds to membrane-bound integrins, inducing their conformational activation and clustering; these activated integrins then recruit molecules to form focal adhesion complexes, which further activate Rho guanine nucleotide exchange factors (Rho GEFs)—the latter promote Rho to switch from its GDP-bound (inactive) state to a GTP-bound (active) state, thus initiating Rho's downstream regulatory functions.

The Rho family of GTPases plays a core regulatory role in this process: RhoA enhances fiber contractility by activating ROCK kinase, providing power for cell migration; Rac1 expands the contact area between cells and the matrix by promoting the formation of lamellipodia, helping cells explore the surrounding environment; Cdc42 determines the direction of cell migration by regulating the extension direction of filopodia(Fig. 5). These three GTPases act synergistically to collectively mediate the directional migration and invasion of tumor cells.⁴⁷ In ovarian cancer, cells perceive the mechanical environment and form a “mechanical memory” effect to maintain the EMT phenotype.⁴⁸ Even if ovarian cancer cells detach from the high-stiffness tumor microenvironment, they can still maintain high metastatic potential, providing favorable conditions for distant tumor metastasis.

Fig. 5.

Mechanical stimulation activates Rho/GTPases (cytoskeleton-regulatory molecules), and its downstream effector RhoA further activates ROCK. On the one hand, ROCK promotes the formation of actin stress fibers; on the other hand, it regulates focal adhesion molecules such as Vinculin. Meanwhile, Rac1 and Cdc42 induce the generation of lamellipodia and filopodia, respectively. These changes in cytoskeletal and adhesion structures inhibit the Hippo signaling pathway (i.e., pathway inactivation). At this point, YAP/TAZ is not phosphorylated, translocates into the nucleus, and binds to the transcription factor TEAD to regulate the expression of related genes, ultimately promoting cell migration. Created with BioGDP.com.³⁵.

2.3. Mechanotransduction in multifaceted tumor progression

Mechanical signals act on the TME in a cell-type-specific manner, and their regulatory effects on stromal cells, immune cells and tumor cells form a complex network, which is an important supplement to the previous single-cell type-focused research on TME mechanoregulation.⁴⁹ Previous studies have focused on the mechanism of regulation on malignant progression of tumor cells (such as migration and invasion) by mechanotransduction signals in the TME. As a key component of the TME, the ECM not only provides physical support for surrounding cells but also its physical properties (e.g., stiffness and fluid stress) play important roles in many biological processes through the synergy of its own functions and other components.^40,50 Different mechanical stimuli in the TME, including elevated matrix stiffness, fluid shear stress, intercellular mechanical tension, ECM topography (oriented collagen fibers) and ECM viscoelasticity, act on their respective key sensors in tumor, immune and stromal cells, and trigger distinct downstream signaling pathways to produce diverse functional outcomes in the TME.¹⁰ Elevated matrix stiffness is sensed by Integrin, Piezo1 and FAK, and activates downstream pathways including Integrin-FAK-RhoA/ROCK, Piezo1-Ca²⁺-PI3K/AKT and YAP/TAZ-TEAD; for tumor cells, this promotes proliferation, migration/invasion, EMT and stemness maintenance, and upregulates MMPs expression to degrade ECM and enhance invasiveness; for immune cells, it induces TAM polarization toward the M2 phenotype and upregulates PD-L1 expression in tumor cells to mediate immune evasion; for stromal cells, it activates CAFs to promote ECM remodeling and further increase matrix stiffness.¹⁰ Fluid shear stress is mainly sensed by Piezo1 and Integrin, and triggers Piezo1-Ca²⁺-AKT/ERK, Integrin-Gq/G11-FAK-NF-κB and Ca²⁺-NF-κB pathways; it regulates circulating tumor cell (CTC) survival in tumor cells (high shear stress induces apoptosis, low shear stress promotes viability),⁵¹ promotes dendritic cell (DC) maturation, upregulates MHC-II and costimulatory molecules (CD80/CD86) and enhances antigen presentation capacity in immune cells,⁵² and modulates endothelial cell alignment and vascular dilation to influence tumor angiogenesis in vascular cells.^22,23 Intercellular mechanical tension is sensed by E-cadherin/cadherin, integrin and actin-myosin cytoskeleton, and activates E-cadherin-RhoA/ROCK and E-cadherin-α-catenin-β-catenin pathways; in tumor cells, this enhances E-cadherin-mediated cell-cell adhesion and collective migration, promotes leader cell formation and local tumor infiltration, coordinates leader-follower cell behavior through ERK activation waves, and regulates nuclear morphology and chromatin remodeling.⁵³ ECM topography (oriented collagen fibers) is sensed by Integrin and DDR1, and activates DDR1-NF-κB-p62-NRF2, Integrin-FAK-RhoA/ROCK and YAP/TAZ pathways; tumor cells directionally migrate along oriented collagen fibers to enhance invasion and metastasis, and DDR1 activation promotes their metabolism and growth⁵⁴; oriented collagen fibers also form a physical barrier to hinder CD4⁺ and CD8⁺ T cell infiltration and induce Treg recruitment to enhance local immune suppression in immune cells,⁵⁵ while promoting collagen cross-linking and oriented arrangement to exacerbate the TME physical barrier in ECM remodeling. ECM viscoelasticity is sensed by Integrin, Piezo channels and YAP/TAZ, and activates Piezo1-Ca²⁺-NF-κB, Integrin-FAK-JAK/STAT and RhoA/ROCK pathways; a viscoelastic/fluid-like matrix promotes TAM migration, drives M2 polarization (CD206⁺, Arg1⁺) and immunosuppressive cytokine secretion (IL-10, TGF-β) and mildly suppresses phagocytosis, while an elastic/solid-like matrix induces M1-like TAM polarization (CD86⁺, iNOS⁺) and pro-inflammatory cytokine secretion (TNF-α, IL-1β), and enhances tumor cell adhesion and proliferation.⁵⁴

2.3.1. Mechanotransduction and tumor microenvironment stiffness

TME stiffness increase is an almost common feature of solid tumors, and this physical property profoundly affects tumor cell behavior through mechanotransduction pathways.

In a stiff matrix environment, mechanical signals are further amplified through integrin, focal adhesion, and FAK pathways, leading to enhanced cytoskeletal contractility and simultaneous activation of the YAP/TAZ signaling pathway.⁵⁶ YAP/TAZ are key effectors of the Hippo pathway ⁵⁷; they are prone to nuclear translocation on stiff matrices. After entering the nucleus, they bind to transcription factors such as TEAD, initiating the expression of genes related to promoting cell proliferation, EMT, and cell stemness, and driving the malignant progression of tumor cells. In breast cancer models, researchers have successfully broken the vicious cycle of “matrix stiffening - mechanical signal activation - YAP/TAZ nuclear translocation - enhanced tumor invasion” by inhibiting lysyl oxidase (LOX) activity⁵⁸ or targeting the YAP/TAZ signaling pathway,⁵⁹significantly reducing tumor invasive potential.

2.3.2. Mechanotransduction and pathways in immune evasion

Mechanical signals not only regulate tumor cells but also profoundly affect the function of immune cells, and synergize with multiple signaling pathways to jointly construct an immune-evasive microenvironment for tumors⁶⁰(Table 3). ECM stiffness, fluid shear stress, intercellular mechanical tension, ECM topography and ECM viscoelasticity, the core mechanical factors of the TME, exert distinct regulatory effects on major immune cell populations in the TME: ECM stiffness induces Tumor-Associated Macrophages (TAMs) toward immunosuppressive M2 polarization and IL-10/TGF-β secretion,⁶¹ impedes Cytotoxic T Lymphocyte (CTL) infiltration and TCR activation while promoting T cell exhaustion, and inhibits NK cell cytotoxicity and activating receptor expression^60,62; Fluid shear stress triggers Piezo1-mediated NET release in tumor-associated neutrophils and drives DC maturation with upregulated MHC-II and costimulatory molecules^63,64; Intercellular mechanical tension promotes Treg enrichment and enhanced immunosuppressive function in the TME⁶³; Oriented ECM collagen fibers form physical barriers to block NK cell infiltration and recruit Tregs for local immune suppression⁶⁵; ECM viscoelasticity modulates TAM polarization, with high viscosity driving M2 polarization and low elasticity inducing M1-like polarization and pro-inflammatory cytokine secretion.⁶⁶ YAP/TAZ functions as a central hub for integrating mechanical signals. It converges signals from the Piezo, integrin, and Hippo pathways, converting extracellular matrix stiffness, cellular tension, and other mechanical cues into transcriptional outputs.^56,67 Meanwhile, YAP/TAZ also possesses independent mechanosensitive regulatory activity; it can directly respond to cellular mechanical status to regulate nuclear localization and transcriptional activity, independently of the classic upstream pathways. Thus, YAP/TAZ exhibits dual properties as both a signaling hub and an autonomous regulator.⁶⁷ In a high-stiffness TME, activated YAP directly binds to the enhancer region of the PD-L1 gene, upregulating the expression level of PD-L1. In BRAF inhibitor-resistant melanoma, high PD-L1 expression inhibits the toxicity of CD8⁺ T cells and reduces the efficiency of anti-tumor immune responses⁶⁸; YAP/TAZ can also promote an immunosuppressive environment by directly upregulating PD-L1 in NSCLC cells, creating an immune-evasive microenvironment for tumor cells.⁶⁹ The integrin-PI3K pathway also plays an important role in tumor immune evasion. ECM stiffness activates the PI3K signaling pathway through integrins, promoting tumor invasion; inhibiting integrin signals can suppress tumor invasion, which implies the role of integrins in ECM stiffness-activated PI3K signaling pathways.⁷⁰ The RhoA-ROCK pathway is also indispensable in mechanical signal-mediated immune evasion. Solid stress in the TME activates the RhoA-ROCK pathway, promoting tumor cells to secrete the chemokine CCL2. CCL2 can recruit monocytic myeloid-derived suppressor cells (M-MDSCs)⁷¹ and inhibit the killing function of natural killer (NK) cells through extracellular vesicles (EVs). In addition, cancer-associated fibroblasts (CAFs) play a key regulatory role in the tumor microenvironment,⁷² on the one hand, they reshape the ECM to increase matrix stiffness and affect the tumor microenvironment; on the other hand, they secrete interleukin 6 (IL-6) to form a cross-activation network with YAP/TAZ, collectively exacerbating tumor immune evasion.

Table 3.

Mechanical regulation mechanisms of immune cells in the tumor microenvironment.

Mechanical Factors	Types of Regulated Immune Cells in TME	Specific Regulatory Effects
ECM Stiffness	Tumor-Associated Macrophages (TAMs)	Induces M2 polarization of macrophages, promotes secretion of IL-10/TGF-β and enhances phagocytosis inhibition of tumor cells.61
	Cytotoxic T Lymphocytes (CTLs)	High-stiffness ECM blocks T cell infiltration, impairs TCR activation, inhibits T cell proliferation/cytotoxic secretion (IFN-γ, perforin), induces Tex and upregulates PD-1/TIM-360
	Natural Killer (NK) Cells	Inhibits NK cell cytotoxicity/killing efficiency, downregulates activating receptors (NKG2D, NKp30) and promotes TGF-β secretion62
Fluid Shear Stress	Tumor-Associated Neutrophils (TANs)	High shear stress: Piezo1 activation, Ca2+ influx, NET release; Low shear stress: reduced NETosis, enhanced vascular adhesion, immune barrier formation63
	Dendritic Cells (DCs)	Promotes DC maturation, upregulates MHC-Ⅱ/CD80/CD86 and enhances antigen presentation.64
Intercellular Mechanical Tension	Regulatory T Cells (Tregs)	Promotes Treg enrichment/expansion, enhances their immunosuppressive function and inhibits effector T cell activation/proliferatio63
ECM Topography	Natural Killer (NK) Cells	Aligned collagen fibers form physical barriers to impede NK infiltration, block effective NK-tumor cell contact and diminish NK-mediated cytotoxicit.65
	Regulatory T Cells (Tregs)	Induces Treg enrichment along oriented fibers to enhance local tumor immunosuppression73
ECM Viscoelasticity	Tumor-Associated Macrophages (TAMs)	High-viscosity: inhibits TAM migration/phagocytosis, promotes M2; low-elasticity: enhances TAM inflammatory secretion, mildly promotes M166

2.4. Mechanotransduction and tumor metastasis

Mechanotransduction drives multiple steps of metastasis and creates favorable conditions for distant tumor metastasis by regulating tumor cell behavior.

In the local invasion stage, increased matrix stiffness in the TME activates the YAP/TAZ-MMPs signaling axis, promoting the expression and secretion of matrix MMPs.⁷⁴ MMPs can degrade basement membrane and ECM components, destroying tissue barriers; at the same time, increased matrix stiffness also increases cytoskeletal tension, accelerating tumor cells to form “invadopodia” and enhancing their ability to penetrate the ECM barrier, laying the foundation for tumor cells to break through the limitations of the primary site.⁷⁵

In the intravasation/extravasation stage, fluid shear stress in blood vessels activates the Piezo1 channel on the surface of tumor cells, mediating Ca²⁺ influx, thereby regulating key Ca²⁺-dependent signaling pathways that control processes related to cancer cell migration, proliferation, and angiogenesis.⁷⁶

In the distant colonization stage, the stiffness of the target organ microenvironment is a key factor determining the colonization efficiency of tumor cells. It is worth noting that there are significant differences in mechanotransduction characteristics among different cancer types. For example, the metastasis process of breast cancer is mainly regulated by tension signals,⁷⁷ while in ovarian cancer and colorectal cancer, the role of fluid shear stress is more significant.^78,79 This difference leads to variations in metastasis mechanisms among different cancers and has become a key gap that needs to be focused on in the current field of tumor mechanobiology research, providing directions for subsequent targeted research.

3. Clinical significance and limitations of the tumor mechanical microenvironment

Aforementioned part discussed that the tumor mechanical microenvironment undergoes corresponding specific changes during tumor progression. Base on this, more sensitive and specific tumor diagnostic methods can be developed, enabling a clearer assessment of the tumor's malignancy and pathological stage. Simultaneously, it allows for artificial intervention in the tumor's living environment to restrict its growth, facilitate the body's own immune response, or improve the prognosis of malignant tumors.

3.1. Innovative tumor diagnostic technologies based on mechanical properties and mechanotransduction molecules

In terms of diagnosis, imaging technologies for abnormal stiffness of the tumor microenvironment have shown broad application prospects. For example, ultrasound elastography⁸⁰ and magnetic resonance elastography⁸¹ can non-invasively detect changes in tumor tissue stiffness, and assist in the early diagnosis and staging of tumors by comparing stiffness differences between tumor and normal tissues. Studies have shown that in various solid tumors such as breast cancer and prostate cancer, the diagnostic accuracy of ultrasound elastography and magnetic resonance elastography is significantly higher than that of traditional imaging examinations.⁸² In addition, the detection of key mechanotransduction molecules (e.g., YAP/TAZ, Piezo1) can also serve as biomarkers for tumor diagnosis and prognosis judgment.

3.2. Novel anti-tumor therapies based on intervening in tumor physical properties

In terms of treatment, targeting the tumor mechanical microenvironment has become a new strategy for tumor treatment, mainly including the following directions: first, reducing tumor matrix stiffness. By inhibiting the activity of LOX family proteins (e.g., using LOX inhibitors), collagen cross-linking is reduced, tumor matrix stiffness is decreased, and the physical properties of the ECM in TME are improved.⁷⁰ This method can not only inhibit the proliferation and invasion of tumor cells but also improve the penetration efficiency of drugs in tumor tissues, enhancing the effect of chemotherapy and targeted therapy.

Firstly, in preclinical models of pancreatic cancer, the combined use of LOX inhibitors and chemotherapeutic drugs significantly improved tumor sensitivity to chemotherapy and prolonged the survival time of model animals. Second, targeting mechanotransduction pathways. Developing specific inhibitors for key mechanotransduction molecules such as integrins, FAK, YAP/TAZ, and RhoA-ROCK to block the transmission of mechanical signals and inhibit the malignant behavior of tumor cells. For example, FAK inhibitors have shown good safety and effectiveness in clinical trials of various solid tumors, which can significantly inhibit tumor growth and metastasis; YAP/TAZ inhibitors have great potential in overcoming tumor drug resistance and can enhance the sensitivity of tumor cells to targeted drugs.¹⁰ Third, combined immunotherapy. Improving the immunosuppressive state of tumors by regulating the mechanical microenvironment. For example, reducing tumor matrix stiffness can reduce the recruitment of immunosuppressive cells (e.g., Tregs, MDSCs) and enhance the infiltration and killing function of CD8⁺ T cells; targeting the YAP/TAZ pathway can downregulate PD-L1 expression and improve the therapeutic effect of immune checkpoint inhibitors.

3.3. Tumor prognosis assessment: from laboratory findings to in vivo applications

In terms of prognosis evaluation, indicators related to the tumor mechanical microenvironment can serve as important prognostic predictors. In addition to the aforementioned association between YAP/TAZ nuclear localization and NSCLC prognosis, in preclinical studies of melanoma, combined administration of YAP inhibitors and PD-1 antibodies significantly enhances the magnitude of anti-tumor immune responses and delays tumor progression. This combinatorial strategy has gradually advanced to the clinical exploration phase, with accumulating relevant cases. In a phase Ⅰ/Ⅱ clinical trial enrolling patients with PD-1-resistant melanoma, 6 subjects with high YAP1 expression in tumor tissues were treated with the YAP inhibitor verteporfin plus PD-1 antibodies. Among these patients, 4 achieved partial response (PR), with liver metastases exhibiting an average shrinkage of 38%. Their progression-free survival (PFS) was prolonged by 6–8 months compared with the period of PD-1 monotherapy. Only mild adverse events (AEs), such as grade 1–2 proteinuria, were observed. Mechanistic analyses demonstrated a significant increase in CD8⁺ T cell infiltration in tumor tissues post-treatment, confirming the role of this combinatorial regimen in reshaping the immune microenvironment.⁸³

Additionally, in clinical studies of NSCLC, advanced patients with positive YAP/TAZ nuclear localization were treated with YAP inhibitors in combination with PD-L1 antibodies. The objective response rate in this cohort was 23% higher than that in the immunotherapy monotherapy group, further validating the clinical value of this combinatorial strategy in tumors with high YAP/TAZ expression.⁶⁹ However, current research is confronted with a major limitation: the inability of in vitro systems to fully recapitulate the mechanical characteristics of tumors in vivo. Such models often fail to capture the spatial heterogeneity of mechanical signals in real tumors, and can hardly simulate dynamic extracellular matrix remodeling, vascular mechanical forces, or mechanical regulation mediated by immune components. This discrepancy between in vitro and in vivo conditions may lead to disconnection between basic research and clinical practice, resulting in biased conclusions. Therefore, it is necessary to critically evaluate existing in vitro models and clarify their limitations. In the future, more physiologically relevant models should be established to promote the clinical translation of related research.

4. Conclusions and future perspectives

From the perspective of tumor microenvironment components, the abnormal remodeling of ECM components such as collagens, proteoglycans, laminins, and fibronectins provides a structural basis for the generation and transmission of mechanical signals; from the perspective of mechanotransduction mechanisms in tumor cells, the four core mechanisms constitute a complete pathway for mechanical signals from perception to conversion; from the perspective of the association between mechanotransduction and TME as well as metastasis, mechanical signals not only promote tumor progression by regulating the behavior of tumor cells themselves but also construct an immune-evasive microenvironment by affecting immune cell functions, and play a driving role in key steps of tumor metastasis such as local invasion, intravasation/extravasation, and distant colonization. In addition, diagnostic technologies, therapeutic strategies, and prognosis evaluation indicators based on the tumor mechanical microenvironment have shown important clinical application value, providing new directions for tumor treatment.

However, the current field of tumor mechanobiology still faces many challenges. First, the mechanical microenvironment exhibits substantial heterogeneity across different cancer types and individual patients. This phenomenon essentially stems from significant physiological differences in the innate matrix composition, collagen density and cross-linking, and baseline tissue stiffness among various organs. Meanwhile, variations in patients’ genetic background, epigenetic status, metabolic profile, inflammatory microenvironment, and tumor driver gene mutation types continuously drive the activation of cancer-associated fibroblasts, extracellular matrix remodeling, and abnormal alterations in matrix stiffness. Coupled with the spatial heterogeneity within tumors, distinct mechanical environments between the invasive front and central necrotic regions, as well as dynamic remodeling driven by disease progression and therapeutic interventions, these factors collectively contribute to the high heterogeneity of the mechanical microenvironment. Such heterogeneity directly leads to marked individual differences in mechanical signal sensing, transduction, and downstream epigenetic regulatory patterns, rendering unified therapies incompatible with the mechanical phenotypes of different patients. Therefore, precisely deciphering the molecular mechanisms underlying this heterogeneity and developing individualized therapeutic strategies represent key issues to be addressed in the future. Second, Mechanical and biochemical signals engage in intricate crosstalk within the tumor microenvironment. Mechanistically, mechanical signals including matrix stiffness, mechanical stretch, and fluid shear stress can directly regulate the activity of growth factor receptors, inflammatory pathways, metabolic pathways, and key enzymes involved in epigenetic modifications via integrins, PIEZO channels, the FAK-RhoA/ROCK pathway, and cytoskeletal remodeling.

Conversely, biochemical signals such as growth factors, cytokines, and hypoxic signals in turn remodel extracellular matrix stiffness, activate mechanosensitive proteins, and alter nuclear lamina and chromatin conformation, forming a multi-layered and multi-node bidirectional regulatory network. At present, the core nodes, upstream and downstream causal relationships, spatiotemporal dynamics, and tissue-specific regulatory patterns through which mechanical and biochemical signals coordinately control tumor proliferation, invasion, metastasis, and immune escape remain incompletely elucidated. The complexity of the signaling network, crosstalk and redundancy among pathways, and the dynamic nature of the in vivo microenvironment further increase the difficulty of mechanistic dissection. Hence, the molecular network underlying the coordinated regulation of tumor progression by these two types of signals still requires in-depth investigation. Finally, although therapeutic strategies targeting the tumor mechanical microenvironment have achieved remarkable progress in preclinical studies, several candidates still face challenges such as unsatisfactory efficacy and prominent side effects in clinical trials. The main reasons include that preclinical models mostly rely on simplified two-dimensional or three-dimensional culture systems and immunodeficient animals, which fail to faithfully recapitulate the complex mechanical microenvironment heterogeneity, mechanical-biochemical signal crosstalk, and intact immune regulatory network in humans. In addition, most candidate drugs target a single mechanorelated pathway and exhibit insufficient capacity to counteract multi-pathway compensatory effects, intratumoral heterogeneity, and dynamic matrix remodeling.

Furthermore, the precise control of tissue distribution, target selectivity, effective concentration, and duration of action in vivo remains challenging. Inhibition of the tumor stroma may simultaneously interfere with the physiological mechanical environment and matrix function in normal tissues, thereby causing off-target effects and toxic side effects. Additionally, differences in mechanical phenotypes and molecular profiles among patients lead to variable drug response rates. Therefore, optimizing drug design and improving targeting specificity and therapeutic efficacy still require continuous exploration. Current clinical cancer treatments are often associated with limitations such as single targeting, high vulnerability to drug resistance due to TME heterogeneity and mechanical-biochemical signal dysregulation, and obvious off-target side effects. To address these clinical translation bottlenecks, therapeutic strategies based on the biomechanical properties of TME have shown unique advantages and broad prospects. The advantages of TME biomechanics-based clinical therapeutic strategies lie in their ability to accurately target core targets, such as abnormal matrix stiffness and mechanical signal transduction pathways, coordinately regulate the mechanical-biochemical signal network, adapt to the individual mechanical phenotype of patients, minimize interference with the physiological mechanical environment of normal tissues, and enhance therapeutic efficacy while reducing side effects. To advance this field, future efforts should focus on using precise detection technologies to clarify the molecular mechanisms of TME mechanical heterogeneity, advance the development of individualized mechanically targeted drugs, optimize multi-pathway combined dosing regimens, achieve synergistic efficacy by combining with immunotherapy and chemoradiotherapy, innovate preclinical models to mimic the real human TME, break through clinical translation bottlenecks, and promote the development of mechanically targeted therapies towards precision, diversification, and synergy.

Research in the field of tumor mechanobiology can be carried out in the following directions: first, strengthening the application of multi-omics technologies. By integrating genomics, transcriptomics, proteomics, and mechanomics data, constructing a more comprehensive regulatory network of the tumor mechanical microenvironment, and in-depth analyzing the molecular mechanisms of mechanical signal regulation of tumor progression; second, developing more precise targeted therapeutic strategies. Based on the mechanobiological characteristics of different tumors, designing drugs with higher specificity and fewer side effects, and exploring the possibility of multi-target combined therapy to overcome tumor drug resistance; third, advancing clinical translation research. Strengthening the combination of basic research and clinical practice, carrying out more clinical trials targeting the tumor mechanical microenvironment, verifying the safety and effectiveness of new therapies, and ultimately transforming basic research results into clinical means to improve patient prognosis. With the continuous advances in this field, the tumor mechanical microenvironment is expected to become an critical breAKThrough in tumor diagnosis and therapy, providing novel theoretical frameworks and technical foundations for achieving precise tumor treatment.

CRediT authorship contribution statement

Xinjie Zhang: Writing – original draft. Jian Sun: Visualization, Methodology. Zhengjian Li: Visualization, Methodology. Shaowei Wang: Writing – review & editing, Supervision.

Ethical approval

This study does not contain any studies with human or animal subjects performed by any of the authors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.