Beyond matrix stiffness: targeting force-induced cancer drug resistance

Abstract

During tumor progression, mechanical abnormalities in the tumor microenvironment (TME) trigger signaling pathways in cells that activate cellular programs, resulting in tumor growth and drug resistance. In this review, we describe mechanisms of action for anti-cancer therapies and mechanotransduction programs that regulate cellular processes, including cell proliferation, apoptosis, survival and phenotype switching. We discuss how the therapeutic response is impacted by the three main mechanical TME abnormalities: high extracellular matrix (ECM) composition and stiffness; interstitial fluid pressure (IFP); and elevated mechanical forces. We also review drugs that normalize these abnormalities or block mechanosensors and mechanotransduction pathways. Finally, we discuss current challenges and perspectives for the development of new strategies targeting mechanically induced drug resistance in the clinic.

Mechanical abnormalities in the TME

In many tumor types, such as breast, colon, and pancreatic cancer, and sarcomas, the TME (see Glossary) is characterized by a desmoplastic reaction, which is directly linked to ECM stiffness [1–3]. ECM stiffness is perhaps the only mechanical aspect of a tumor that can be assessed by clinicians and patients via palpation and can lead to clinical detection of many solid tumors (e.g., in breast or prostate cancer) [1]. As cancer cells proliferate rapidly within the stiff ECM of a solid tumor at the expense of the host tissue, a reciprocal compressive force is applied by the host tissue to resist tumor expansion [2,3]. This leads to accumulation of mechanical stress (i.e., force per unit area) within the tumor [2,4], which is compressive in the tumor interior (i.e., tends to reduce the size of an object). At the interface between the tumor and normal tissue, the stress is tensile (i.e., tends to increase the size of an object) [2,3]. At the cellular level, mechanical compression affects cell cycle progression, migration, invasion, and proliferation [4–8]. At the tissue level, increased compressive forces lead to blood and lymphatic vessel compaction that can impair the transport of nutrients, oxygen (resulting in hypoxia), therapeutic agents, and cells inside the tumor [3]. Hypoxic environments can further fuel tumor progression via induction of the cancer stem cell (CSC) phenotype, epithelial-to-mesenchymal transition (EMT), angiogenesis, inflammation, fibrosis, and immunosuppression [9]. The dense ECM can also act as a physical barrier to immune cell infiltration, contributing to the formation of an immune-excluded phenotype [10,11]. In response to a hypoxic environment, new blood vessels are formed to supply oxygen to the rapidly growing tumor. However, tumor-induced angiogenesis can lead to immature vessels, characterized by reduced pericyte coverage, detachment or loss of the basement membrane, and large openings between endothelial cells, causing vessel hyperpermeability (i.e., leakiness) and transport of large amounts of fluid from the vascular to the extracellular space of the tumor [12,13]. In turn, this results in uniformly elevated IFP, washing drugs out of the tumor [12,13].

Sensing of mechanical cues arising from these TME abnormalities activates mechanotransduction-triggered pathways, which regulate multiple cellular phenotypes (e.g., proliferation, invasion, and survival). Mechanotransduction is essential to tissue homeostasis; however, cancer cells co-opt these aberrant TME signals to activate downstream signaling pathways that lead to abnormal cellular responses, including cell cycle arrest, disruption of apoptosis, immune evasion, and improved DNA repair. Current strategies to overcome therapy resistance have had limited success, because they focus on cell-intrinsic pathways and genetic alterations without considering the impact of the ECM and TME mechanical cues that enable cancer cells to evade death. To this end, a better understanding of the mechanisms involved in mechanical cue sensing and how they regulate cellular programs that determine drug sensitivity will advance the development of novel therapies that target mechanically induced therapy resistance.

In this review, we describe the mechanisms of action and resistance of anti-cancer therapeutic strategies, with a focus on chemotherapy. In addition, we highlight how the biomechanical abnormalities of TME lead to drug resistance, illustrating the crosstalk with mechanical signaling pathways. We next focus on two different therapeutic approaches that either restore/normalize the TME or target components of mechanical signaling, and classify them based on whether they have been tested preclinically, whether they are currently in clinical trials, or are FDA approved. We also discuss opportunities and challenges in mechanoresistance, as well as implementing these mechanical-targeted strategies in the clinic.

Mechanisms of chemotherapy action and resistance

Chemotherapies have diverse mechanisms of action to induce cytotoxic and cytostatic effects on cancer cells. However, response to chemotherapy critically depends on intrinsic mechanisms of chemoresistance that upregulate DNA repair, drug efflux, prosurvival signaling, and cell cycle arrest. The mechanisms of action and resistance of the most common chemotherapies (platinum alkylating agents, topoisomerase inhibitors, mitotic inhibitors, and antimetabolites) that have also been linked with mechanical-induced signaling are presented below. For more details about these mechanisms, we refer readers to several comprehensive reviews [14,15].

Chemotherapeutics can be broadly classified into several categories with diverse mechanisms of action. Platinum alkylating agents, such as cisplatin and carboplatin, covalently bind to DNA to form inter- and intrastrand adducts that disrupt DNA repair, induce DNA damage, and activate proapoptotic genes [16]. Topoisomerase II inhibitors, such as doxorubicin, etoposide, and mitoxantrone, stabilize the topoisomerase II cleavage complex during DNA replication, resulting in DNA single- and double-strand breaks that induce apoptosis [17]. Paclitaxel and docetaxel are microtubule inhibitors that cause cell cycle arrest due to their disruption of microtubule formation and disassembly, which are necessary for the formation of mitotic asters and the kinetochore during cell division [18]. Antimetabolites, such as 5-fluorouracil (5-FU) and gemcitabine, incorporate into RNA or DNA instead of their normal nucleotide counterpart. 5-FU incorporates into DNA to inhibit DNA and RNA synthesis [19], whereas gemcitabine also induces DNA damage to prevent DNA replication [20].

Cancer cells can evade the effects of chemotherapy through several resistance mechanisms. Here, we discuss the preclinical and clinical evidence for mechanisms that are relevant to mechanoresistance. For example, heightened DNA repair capacity can correct chemotherapy-induced lesions to prevent cell death. Greater expression of genes associated with DNA repair signaling pathways is correlated with increased resistance to cisplatin in ovarian and lung cancer, and copy number amplification of these genes is associated with reduced overall survival for several cancers [21]. ATP-binding cassette (ABC) family multidrug efflux pumps sequester or remove drugs from the cell. P-glycoprotein (ABCB1) is the most prominent efflux pump [22] and is associated with poor response to many chemotherapies and overall survival in patients with ovarian cancer [23]. A greater ratio of proapoptotic proteins to antiapoptotic proteins triggers apoptosis [24]. The expression of these proteins is regulated by many pathways, such as PI3K/Akt and p38-MAPK signaling [24,25]. Increased Akt expression is associated with paclitaxel and cisplatin resistance in ovarian cancer cells and doxorubicin resistance in gastric cancer [26–28]. Additionally, greater expression of antiapoptotic proteins BCL2L1 and MCL1 is associated with worse survival outcomes in patients with ovarian cancer [29]. Cell cycle checkpoints regulate progression through the cell cycle. Cancer cell sensitivity to certain chemotherapies (e.g., platinum based and taxanes) are cell cycle specific; thus, arrested cells may be less sensitive to the effects of chemotherapy [30,31]. Patients with cisplatin-treated non-small-cell lung cancer expressing high p27^Kip1, which prevents cell cycle progression from G₁ to S phase, experienced poorer overall survival compared with patients with low p27^Kip1 expression [32]. These pathways mediating chemoresistance can be activated in a cell-intrinsic manner (i.e., genetic alterations) or in a cell-extrinsic manner via environment-mediated chemoresistance. We discuss this intricate connection between therapeutic resistance and mechanical forces/mechanotransduction signaling mechanisms in the next section.

Mechanical signaling crosstalk with drug resistance mechanisms

Mechanical cues can regulate response to chemotherapy by activating chemoresistance programs that are associated with cell states, such as EMT and cancer stemness. Cells that have undergone EMT exhibit increased expression of drug efflux pump genes, resistance to apoptosis, anchorage-independent growth (anoikis), and stem cell characteristics [33]. Furthermore, CSCs exhibit chemoresistance due to their slow proliferation, high prosurvival programs via Akt/PKB signaling, greater drug efflux, and efficient DNA damage repair [34]. Here, we discuss the role of these cell states on chemoresistance and how mechanical abnormalities impact chemotherapy response (see Box 1 for mechanosensing and Figure 1 for shared chemotherapy with mechanotransduction signaling responses). Studies related to mechanically induced chemoresistance are summarized in Table 1. We also discuss several studies that provide evidence for the mechanisms of mechanically induced immunotherapy and targeted therapy resistance (see Box 2).

Box 1. Mechanisms of sensing and transmitting mechanical cues.

By forming a physical boundary between intracellular components and the extracellular environment, the cellular membrane is a key interface mediating response to mechanical stimuli. Cellular membranes contain a variety of mechanosensitive proteins that translate extracellular mechanical cues into intracellular signals (see Figure 1 in main text). Among the most well-characterized mechanosensitive proteins are integrins, found to be overexpressed in solid tumors, including breast, pancreatic, and lung carcinomas [134]. Upon activation, integrins recruit the adhesion plaque protein talin, creating mature focal adhesions [135]. Focal adhesions physically connect integrins to the cytoskeleton by their interaction with actin linker molecules, inducing the formation of actin stress fibers in the cytoplasm [135]. In addition, focal adhesions interact with signaling molecules, including focal adhesion kinase (FAK), which in turn presents a docking site for the nonreceptor tyrosine kinase Src [135]. Downstream, the FAK/Src complex activates the NF-κB, Hippo, MAPK, and PI3K signaling pathways, which can also interact with several receptor tyrosine kinase (RTK) and growth factor receptors (GFRs), including the epidermal (EGFR), insulin (IGFR), and transforming growth factor beta (TGFβR) receptors [136].

In addition to integrins, mechanically activated ion channels, including Piezo and transient receptor potential (TRP), and GPCRs are cell surface receptors of mechanical signals. Ion channels are nonselective cation channels responding with channel gating to mechanical stress [137] or indirectly by integrins or other scaffolding elements, such as ECM components [94]. Ion channels sense mechanical cues in the tissue microenvironment and cooperate with focal adhesions to regulate cellular responses. In breast cancer cells, upon application of mechanical stress, activated mechanosensitive ion channels were detected in 54% of cell membrane patches, leading to higher metastatic potential [138]. Similarly, GPCRs interact with nearby G proteins on the plasma membrane and cooperate with integrins and GFRs to regulate cytoskeletal rearrangements and cell shape changes [139]. Studies have shown that GPCRs are overexpressed in a variety of cancer types, contributing to cell proliferation in response to ligands or mechanical cues [140].

Mechanical sensing and mechanotransduction pathways are also necessary for other cellular components of the TME, including stromal and immune cells. The cellular responses mediated in these cell types can also affect tumor progression and resistance to therapy, and are extensively reviewed elsewhere [141,142].

Figure 1. Shared signaling responses induced by mechanical cues and mechanisms of chemoresistance in cancer cells.

Mechanical cues that are present in the tumor microenvironment [TME; i.e., mechanical forces or extracellular matrix (ECM) stiffness] are sensed by: (i) membrane receptors, such as integrins; (ii) G protein-coupled receptors (GPCRs); and (iii) ion channels. At the same time, mechanosensors regulate the activity of GFRs, together leading to the activation of major signaling pathways, including RAS, PI3K, FAK, and Hippo. These pathways interact with each other to promote cytoskeleton rearrangement, which is mainly facilitated by the activation of ROCK/Rho, RAC/cdc42, and myosin II, translocation of transcription factors into the nucleus that directly regulate gene expression, and eventually regulation of epithelial-to-mesenchymal transition (EMT) and cancer stemness. Altogether, these mechanically driven responses lead to four main chemoresistance mechanisms: activation of (i) drug efflux and (ii) DNA repair, and alterations in (iii) cell cycle and (iv) apoptosis. Faded structures represent known mechanically regulated pathways in normal cells, while structures shown in full color represent those structures shown to be regulated by mechanical cues in cancer cells. Abbreviation: CSC, cancer stem cell. Created with BioRender.com.

Table 1.

Studies that identified mechanically induced chemoresistance mechanisms

Experimental setup (simulated mechanical cue of TME)	Cancer type	Mechanotransduction mechanism	Chemoresistance mechanism	Resistance to	Refs
ECM composition and stiffness
Fibronectin (FN)-coated substrate (ECM component, 2D)	Ovarian; breast	FN/PI3K/Akt2/survivin axis activation	Inhibition of apoptosisa	Docetaxel (microtubule inhibitor)	[37]
	Myeloma	FN/β1-integrins/p27kip1 axis activation	Inhibition of DNA synthesis → cell cycle arrestb	Etoposide VP-16 (topoisomerase II inhibitor)	[38,39]
	Breast	FN/β1-integrins/PI3K/Akt axis activation	Inhibition of cytochrome c release/sustained expression of Bcl-2 proteina	Paclitaxel (microtubule inhibitor)	[40]
Collagen I and FN substrate (ECM components, 2D)	Breast	Collagen I–integrin activation	Activation of integrin-induced ABC efflux transporter (CAM-DR)c	Cisplatin (platinum alkylating agent), doxorubicin, mitoxantrone (topoisomerase II inhibitors)	[46]
Collagen I substrate (ECM component, 2D)	Jurkat and HSB2 leukemic T cells	Collagen I–integrin/Erk activation	Upregulation of ABC drug transporters → drug effluxc	Doxorubicin (topoisomerase II inhibitor)	[48]
3D collagen I matrix (ECM component)	Pancreatic	Increased membrane type 1 matrix metalloproteinase (MT1-MMP)/ERK1/2 phosphorylation → increased HMGA2 expression → chromatin remodeling	Activation of checkpoint kinase 1 (Chk1) and cell cycle regulationb	Gemcitabine (antimetabolite)	[145]
	Ovarian	Activation of TGFβ and Wnt signaling, increased MMP, fibronectin, vimentin, N-cadherin, Slug, Snail, and integrin expression	Activation of TGFβ-, Wnt-induced EMTd → decreased growth ratese	Carboplatin (platinum alkylating agent), 5-FU (antimetabolite), paclitaxel (microtubule inhibitor)	[55]
Increased stiffness of polyacrylamide gels (ECM stiffness)	Pancreatic	Increased vimentin and decreased E-cadherin expression; nuclear translocation of β-catenin, YAP and TAZ	Induction of EMTd,e	Paclitaxel (microtubule inhibitor)	[56]
Increased stiffness of methylcellulose matrix (ECM stiffness)	Pancreatic	Alterations in gene expression (drug transporter, collagen, fibronectin, and metabolic genes)	Increase in drug efflux, CAM-DR, alterations in cell metabolismc	Multiple drugs	[51]
Increased stiffness of polyacrylamide hydrogel (ECM stiffness)	Breast	Inactivation of Hippo pathway protein; Merlin/MST/LATS/ILK-induced activation of YAP	Increase in cell survivala, upregulation of ABCB1 expression → drug effluxc	Doxorubicin (topoisomerase II inhibitor)	[47]
HA-containing Matrigel matrix (ECM stiffness)	CSC of head and neck squamous cell carcinoma	HA-CD44v3 activation → increased miR-302 expression	Upregulation of prosurvival proteins (e.g., AP1) and CSC markers (e.g., Oct-4), cancer cell self-renewal, stemness, and clonal formatione	Cisplatin (platinum alkylating agent)	[69]
Increased stiffness of alginate beads (ECM stiffness)	Hepatocellular	Not identified	Alterations in expression of ABC drug transporters → drug effluxc; upregulation of antiapoptotic programa	5-FU (antimetabolite), cisplatin (platinum alkylating agent), paclitaxel (microtubule inhibitor)	[52]
Increased stiffness of collagen I matrix (ECM stiffness)	Hepatocellular	Not identified	Alterations in expression of drug metabolism enzyme CYP3A4c,d	Doxorubicin (topoisomerase II inhibitor)	[146]
	Colorectal	FAK/Wnt signaling activation	Induction of CSC → inhibition of cell growth	5-FU-based chemotherapies (antimetabolite)	[66]
Increased stiffness of alginate hydrogels (ECM stiffness)	Breast	Nuclear translocation of YAP, decreased E-cadherin expression	Induction of EMTd,e	Doxorubicin (topoisomerase II inhibitor)	[57]
Increased stiffness of polyacrylamide hydrogels (ECM stiffness)	Breast	Regulation through MAP4K4/6/7 kinases/RNF8-mediated ubiquitin signaling at DNA double-stranded breaks (DSBs)	Induction of DNA DSB repairf	Etoposide (topoisomerase II inhibitor), cisplatin (platinum alkylating agent)	[42]
Mechanical forces (compression/shear)
Flow-induced shear stress using 3D tumoroid flow system (ECM stiffness, IFP-induced shear stress)	Breast	Upregulation of Snail/MMP14 and downregulation of E-cadherin/vimentin	EMTd,e; inhibition of cell viability/proliferation	Doxorubicin (topoisomerase II inhibitor)	[91]
Hydrostatic pressure generated using microfluidic platform (increased fluid pressure)	Breast	Not identified	Modulation of expression of drug transporter gene ABCC1 → drug effluxc	Doxorubicin (topoisomerase II inhibitor)	[88]
3D-matrigel flow culture (increased flow-induced shear stress)	Ovarian	EGFR/MEK/ERK signaling activation	–	Carboplatin (platinum alkylating agent)	[48]
PDMS microfluidic channel (increased flow-induced shear stress)	Ovarian	miR-199a-3p/PI3K/Akt/ABCG2 drug transporter activation	Alterations in drug effluxc; EMTd,e; CSCd	Cisplatin (platinum alkylating agent); paclitaxel (microtubule inhibitor)	[89]
3D hydrogel embedded in bioreactor that applies shear stress (increased flow-induced shear stress)	Breast	Upregulation of PLAU expression and increase in uPA activity	Increase in cell proliferation and invasionb	Paclitaxel (microtubule inhibitor)	[90]
Compression bioreactor system; 3D (compression-induced mechanical stress)	Ovarian	CDC42 upregulation	Upregulation of cdc42, Oct4 → EMT and CSCe; increase in ABC drug transporters expression → drug effluxc; inhibition of apoptotic cell deatha	Paclitaxel (microtubule inhibitor) and carboplatin (platinum alkylating agent)	[99]
Spheroids embedded in 3D agarose matrix (compression-induced mechanical stress)	Pancreatic	Inactivation of ERK	Decrease in cell proliferationb	Gemcitabine (antimetabolite)	[98]

^aApoptosis/cell survival.

^bCell cycle.

^cDrug efflux.

^dThe exact cellular process that leads to chemoresistance has not been specified.

^eEMT/CSC.

^fDNA repair.

Box 2. Mechanisms of targeted and immunotherapies.

Mechanism of action and resistance of targeted therapies

Targeted therapies present a personalized approach for treating cancer based on genetic alterations in individual patients. Prominent examples include agents targeting the EGFR, RAF/MEK, and AKT signaling pathways, which are aberrantly activated in cancer cells to drive proliferation and survival, and evade apoptosis. In addition, VEGF inhibitors represent another class of targeted therapies that are used to interfere with the formation of new blood vessels and limit tumor growth. Despite the initial benefits offered by targeted therapies, their effectiveness often diminishes over time due to new genetic alterations or reactivation of the target molecule, upstream or downstream signaling activation, epigenetic changes, by-pass signaling changes, and cancer cell–ECM interactions that enable cell survival under therapy (extensively reviewed in [143]).

Immunotherapy mechanisms of action and resistance

Distinct from chemotherapy, which impacts all cell types, immunotherapy harnesses the immune system to kill cancer cells. Immune checkpoint inhibitors (ICIs) represent one of the most successful immunotherapies in the clinic. Their mode of action involves inhibition of checkpoint molecules that act as ‘brakes’ on the immune system, with PD-1 being a widely targeted molecule. The PD-1 receptor, which is expressed on T cells, recognizes PD-L1 ligand on cancer cells and prevents T cell-mediated killing. By blocking PD-1/PD-L1 recognition, the cytotoxicity of T cells is restored, allowing them to recognize and attack cancer cells more effectively [11]. However, despite the significant advances in ICI as a cancer treatment, many patients do not respond to this approach. ICI resistance is usually attributed to the absence of tumor-infiltrating effector immune cells and TME properties [80].

Although cell-intrinsic mechanisms of drug resistance have been extensively studied, recent research highlights the role of extracellular cues, including mechanical forces in the TME, in promoting drug resistance [144]. Therefore, investigating the coupling between mechanical signaling and therapy resistance mechanisms is crucial to developing novel strategies that can enhance the effectiveness of targeted and immune-directed therapies.

ECM mechanisms of therapy resistance

Adhesion of cancer cells to ECM components, such as collagen and fibronectin, or their growth in a stiff matrix, drives resistance to chemotherapy [35]. This ECM-mediated chemoresistance mechanism is primarily based on activation of integrin signaling, which results in overexpression of prosurvival and antiapoptotic proteins, cell cycle arrest, modulation of drug efflux, and a phenotype switching of cancer cells (e.g., EMT or cancer stemness), which is referred to as cell adhesion-mediated drug resistance (CAM-DR) [36]. For example, ovarian and breast cancer cells exhibited docetaxel resistance through PI3K/Akt/survivin axis-induced inhibition of apoptosis [37], while, in myeloma cells, fibronectin-induced integrin activation led to cell cycle arrest and etoposide resistance [38,39]. In breast cancer cells, it has been also shown that fibronectin-induced integrin activation inhibited the release of cytochrome c and the expression of the antiapoptotic protein Bcl-2 through the PI3K/Akt signaling axis, resulting in inhibition of apoptosis triggered by microtubule-directed chemotherapeutic agents [40]. Another study demonstrated that gemcitabine resistance can be induced by activation of checkpoint kinase 1 (CHK1) and subsequent cell cycle regulation through the MMP/ERK1/2 signaling axis in pancreatic cancer cells growing in a 3D collagen matrix [41]. In breast cancer cells, increased stiffness of fibronectin-coated substrates was also shown to facilitate DNA repair of double-stranded DNA breaks, impairing the efficacy of several drugs, including etoposide and cisplatin. This mechanism was regulated by MAP4K4/6/7 kinase and subsequent ubiquitin phosphorylation, which recruits H2AX at DNA-damaged sites to activate DNA repair mechanisms [42]. Transcriptomic-based analysis in the clinic of patients with breast and gastric cancer also provides evidence for the role of ECM constituents, such as fibronectin and collagen type I, and poor treatment responses [43–45].

Recent studies showed that ECM stiffness has also a significant effect on the activity of ABC drug transporters that determine intracellular drug uptake [46]. When the ECM is stiff, ABC transporters are less active and less effective at removing drugs from cells. Conversely, when the ECM is more compliant or soft, ABC transporters are more active, which can enhance drug clearance and improve treatment outcomes [47]. Breast cancer cells growing on collagen I and fibronectin substrates exhibited resistance to cisplatin, doxorubicin, and mitoxantrone through elevated expression of ABC efflux transporters [46,47]. In a similar manner, Jurkat and HSB2 leukemic cells grown on collagen substrate acquired resistance to doxorubicin through integrin/Erk signaling activation and ABCC1 drug transporter upregulation [48]. Clinically, matrix stiffness has been measured directly through elastography or inferred through lysyl hydroxylase (LOX) expression, and is correlated with poorer pathological complete response and relapse-free survival [49,50]. Alterations in the expression of ABC drug transporter genes have also been observed in pancreatic [51] and hepatocellular carcinoma cells [52] grown on substrates of increased stiffness, which led to resistance to 5-FU, cisplatin, and paclitaxel. These findings highlight the importance of ABC transporters as key regulators of ECM-induced chemoresistance, and should be a focus of future research and therapeutic development efforts.

Cell adhesion to collagen type I and fibronectin upregulates the expression of mesenchymal markers, including Slug, Snail, and vimentin, and downregulates E-cadherin to promote EMT transition through the activation of Src and the ERK/MAPK pathway [53] or the focal adhesion kinase (FAK)/Src/β-catenin signaling axis [54]. In line with this, ovarian tumor spheroids embedded in type I collagen-rich ECM showed increased invasive abilities and EMT, which subsequently led to chemoresistance, although the exact mechanism was unclear [55]. In pancreatic cancer cells, increased stiffness upregulated EMT markers and promoted the nuclear localization of β-catenin, YAP, and TAZ mechanoresponsive transcriptional regulators, leading to paclitaxel resistance [56]. In breast cancer cells, dynamic ECM stiffening induced doxorubicin resistance through YAP nuclear translocation and an EMT state [57].

ECM signals, including transforming growth factor beta (TGFβ) release in the TME, may also induce a CSC phenotype by activation of EMT in cancer cells [58,59] and maintenance of cancer cell dormancy [60]. In colorectal and breast cancer, type I collagen promotes the expression of CSC markers [61–63], while in squamous cell carcinoma and breast cancer, it increases CSC enrichment through activation of FAK signaling [64] and Akt/mTOR/YAP signaling [65], respectively. In a similar manner, it was shown that cancer cells reversed their mesenchymal phenotype, reduced their proliferation rate, and acquired chemoresistance when co-cultured with hepatic stem cells in organoids embedded in dense collagen type I matrices [66]. Similar to collagen type I, hyaluronic acid (HA)-rich ECM also provides a favorable environment for the maintenance of CSCs. This is evidenced by studies showing that hyaluronan synthase (HAS2) is necessary for the metastatic ability of breast CSCs [67], HA accumulation increased CSC enrichment in breast tumor models through Twist- and TGFβ-snail signaling axis [68], and the interaction of HA with CD44 in head and neck squamous CSC promoted their self-renewal and cisplatin resistance [69]. Altogether, these studies support that ECM composition has an important role in CSC survival/growth, which, in turn, leads to chemoresistance and the increased metastatic potential of cancer cells.

In addition to chemoresistance, mechanical cues also affect the efficacy of targeted therapies. Indeed, patient-derived glioblastoma cells grown in stiff hydrogels comprising HA, showed reduced response to the FDA-approved epidermal growth factor receptor (EGFR) inhibitor erlotinib [70]. Furthermore, collagen-driven activation of integrin signaling in lung cancer cells also conferred resistance to multiple EGFR inhibitors [71]. Increased ECM stiffness in colorectal liver metastasis reduced the efficacy of the antiangiogenic agent bevacizumab [72], as well as the efficacy of the Raf inhibitor sorafenib in breast cancer cells through the β_1-integrin/JNK signaling axis [73]. Finally, ECM stiffness induced resistance of breast cancer cells to the HER2 inhibitor lapatinib through YAP/TAZ transcriptional activity [74], and to BRAF inhibitors in melanoma cells [75].

A dense ECM constitutes a physical barrier to the diffusion of immune checkpoint therapies [α-programmed cell death protein 1 (α-PD-1)/programmed death-ligand 1 (PDL-1)] and T cell infiltration, which limit cancer cell cytotoxicity [76,77]. Trafficking of T cells through dense collagen fibers can result in nuclear damage, impaired motility, and eventually cell death [78]. Increased matrix stiffness has also been shown to upregulate PD-L1 expression in cancer cells that was linked with YAP/TAZ nuclear translocation [79]. Other cellular components of the immune system are also affected by increased matrix stiffness, including dendritic cells, natural killer cells, and macrophages, which exhibit chemotaxis toward cancer cells and determine the balance between an immunostimulatory versus immunosuppressive microenvironment [80]. For example, when lysyl oxidases were used to target ECM remodeling and stiffness, T cell migration improved with increased efficacy of anti-PD-1 blockade [81]. In line with this, restoring TME abnormalities using the mechanotherapeutic tranilast, combined with immune checkpoint inhibition, increased T cell infiltration in TME, and immunological memory in mice bearing immunotherapy-resistant breast cancer [82].

Force-induced therapy resistance

In solid tumors, ECM stiffness is mainly determined by ECM composition and organization, while mechanical/physical forces are exerted during tumor growth. At the tissue level, the compressed and abnormal tumor blood vessels along with the increased deposition of matrix components in the tumor ECM, hinder the delivery of therapeutic agents to the tumor interior [83,84]. At the same time, the elevated intratumoral IFP, which drops to normal values at the tumor periphery, induces outward fluid flow from the tumor interior to the surrounding tissue, which washes drugs out of the tumor [83,85,86]. At the cellular level, while the effect of ECM stiffness on therapy resistance is actively being investigated, studies focused on force-induced drug resistance remain limited. Here, we provide evidence for how mechanical forces compromise the effectiveness of anti-cancer therapies, illustrating the molecular mechanisms identified so far.

Elevated IFP-induced shear stress and therapy resistance

Within the TME, the accumulation of interstitial fluid due to blood vessel hyperpermeability and lymphatic dysfunction leads to IFP, which hinders drug delivery. High IFP values have been also measured in patients with cervical cancer, and are considered an independent poor prognostic factor for tumor recurrence after radiotherapy [87]. However, the direct effects of IFP on cancer cell chemosensitivity remain poorly understood. Breast cancer cells exposed to hydrostatic pressure exhibited poor response to doxorubicin, which was mediated by upregulation of the ABCC1 drug transporter and reduced intracellular doxorubicin concentration [88]. High IFP also generates shear forces on tumor cells via fluid flow. In ovarian cancer spheroids, shear stress increased the expression of EMT markers, ABCC1 drug transporters, and cancer stemness markers, with reduced sensitivity to cisplatin and paclitaxel [89]. Additionally, another study showed that adherent ovarian cancer cells exhibited a poor response to carboplatin following exposure to shear stress through activation of EGFR-driven MEK and ERK prosurvival signaling [48]. Consistent with these results, breast cancer cells exposed to fluid shear stress also exhibited increased motility and resistance to paclitaxel [90]. When shear stress was applied on breast cancer cells embedded in a 3D collagen matrix, it was shown to induce the expression of EMT markers and result in poor response to doxorubicin across multiple cell lines [91]. Importantly, transcriptomic analysis of patients with triple-negative breast cancer demonstrated that the expression of chemoresistance-related genes was correlated with a shear stress-induced gene expression profile [92].

Although little is known about IFP- and shear stress-induced targeted therapy resistance, it was shown that shear stress enhanced the resistance of sarcoma cells to insulin growth factor-1 receptor blockade by dalotuzumab [93]. IFP also impairs the delivery of immunotherapies to the tumor interior, and induces flow of tumor- and stromal cell-derived immunosuppressive exosomes toward the tumor boundary. These exosomes may recruit immunosuppressive immune cells to fuel tumor progression [94].

Compression-induced mechanical stress and therapy resistance

Mechanical forces result not only internally from TME structural remodeling, but also externally from the cancer–host tissue reactive boundary due to tumor expansion. These compressive mechanical cues have been linked with alterations in the proliferation rate, higher metastatic potential, and survival of cancer cells [4–8,95,96]. However, their role in chemoresistance remains elusive [97]. It has been shown that compressive stress results in mitotic arrest through inhibition of the bipolar spindle assembly, which eventually impairs proliferation [5]. Given that most common chemotherapeutic agents target proliferating cells, this compression-induced mechanism may limit therapeutic responses. Another study that combined mathematical modeling with experiments showed that high compressive stress generated during tumor spheroid growth in a confined agarose matrix led to decreased cell proliferation and response to gemcitabine, which targets actively proliferating cancer cells [98]. Mechanical compression applied in a 3D setting also resulted in ovarian cancer invasion and chemoresistance through upregulation of CDC42 expression; however, the underlying mechanotransduction mechanism is not yet defined [99]. Mechanical stresses were also shown to induce immunotherapy resistance through activation of the PI3K/Akt pathway in tumor cells, which blunts T cell-induced apoptosis, increases PD-L1 expression, and promotes the recruitment of immunosuppressive immune cells (e.g., Tregs) [61]. The effect of compression-induced mechanical stress on the efficacy of targeted therapies remains to be explored.

Strategies to target biomechanical abnormalities in the TME and mechanically induced signaling pathways

Mechanotransduction pathways and the mechanically remodeled TME can be therapeutically targeted by blocking mechanical signaling components and normalizing the TME (Figure 2, Key figure). In Table 2, we list drugs that have been tested in clinical trials in combination with the standard-of-care, as well as the outcome of the clinical trial.

Figure 2. Drugs that target mechanosensors, growth factor receptors, mechanotransduction pathways, or components of the extracellular matrix (ECM).

Grey boxes indicate drugs that are in preclinical studies; blue boxes include drugs that are being tested in clinical trials; and green boxes include drugs that are FDA approved for the treatment of patients with cancer. Names in parentheses correspond to the target of each drug. Abbreviations: CAFs, cancer-associated fibroblasts; GFRs, growth factor receptors; HA, hyaluronic acid; MMPs, matrix metalloproteinases. Created with BioRender.com.

Table 2.

Representative clinical trials of agents targeting the mechanical aspects of TME

Target	Agent	Phase	Cancer type	Combined with	Clinicaltrial.gov ID
Mechanosensors/mechanotransduction
Integrin	Cilengitide	2	Squamous cell carcinoma	5-FU + cisplatin + cetuximab (EGFR	NCT00705016 a
				inhibitor) (PFE)
		1	Breast	Paclitaxel	NCT01276496 b
		2	Lung	Cisplatin, gemcitabine, cetuximab	NCT00842712
		2	Glioblastoma	Temozolomide	NCT00085254
		3	Glioblastoma	Temozolomide	NCT00689221 a
	Volociximab (M200)	1b	Lung	Carboplatin, paclitaxel, bevacizumab (anti-VEGF)	NCT00666692b, NCT00654758b
		2	Lung	Erlotinib	NCT00278187 b
		2	Pancreatic	Gemcitabine	NCT00401570 b
		1/2	Ovarian	Doxorubicin	NCT00635193 a
		2	Melanoma	Dacarbazine	NCT00099970 b
	E7820	1&2	Colon, rectal	Irinotecan versus FOLFIRI	NCT01347645 b
		2	Colorectal	Cetuximab	NCT00309179 b
		1/2	Colorectal	FOLFIRI versus bevacizumab	NCT01133990 b
	Abituzumab (EMD 525797)	1/2	Colorectal	Cetuximab + irinotecan	NCT01008475 a
		2	Colorectal	Cetuximab + FOLFIRI	NCT03688230 a
	ATN-161	1/2	Glioma	Carboplatin	NCT00352313 b
	Vitaxin	2	Prostate	Docetaxel	NCT00072930 b
		1/2	Renal	Bevacizumab	NCT00684996 a
		2	Melanoma	Dacarbazine	NCT00066196 c
FAK	Defactinib	1	Lung	VS-6766 (RAF/MEK inhibitor)	NCT03875820 d
		1/1b	Ovarian	Paclitaxel	NCT01778803 b
		1/2	Ovarian	Paclitaxel, carboplatin	NCT03287271 d
		1	Pancreatic	Gemcitabine + pembrolizumab (immunotherapy)	NCT02546531 b
	GSK2256098	2	Pancreatic	Trametinib (MEK inhibitor)	NCT02428270 b
		2	Meningiomas	Vismodegib (Hedgehog inhibitor), capivasertib (Akt inhibitor), abemaciclib (CDK inhibitor)	NCT02523014 d
	VS-4718	1	Pancreatic	Nab-paclitaxel + gemcitabine	NCT02651727 e
	IN10018	1	Gastric	Docetaxel	NCT05327231 d
		1	Melanoma	Cobimetinib (MEK inhibitor)	NCT04109456 d
β-catenin	PRI-724	2	Colorectal	Bevacizumab + oxaliplatin + leucovorin + fluorouracil	NCT02413853 e
		1	Pancreatic	Gemcitabine	NCT01764477 b
Tyrosine kinases	Dasatinib (>300 trials)	1/2	Lung	Erlotinib	NCT00826449 b
		2	Pancreatic	Mfolfox6	NCT01652976 a
		FDA approved	Myeloid leukemia	–	–
CXC4R (GPCR)	TG-0054 (burixafor)	2	Multiple myeloma, lymphoma	G-CSF	NCT02104427 b
TME normalization
Stromal components	S-3304 (MMP)	1/2	Lung	Chemo-irradiation	NCT00078390 b
	PEGPH20 (HA) (discontinued)	1/2	Pancreatic	Gemcitabine/nab-paclitaxel + dexamethasone + gemcitabine + enoxaparin	NCT01453153c, NCT01839487c
		3	Pancreatic	Gemcitabine + nab-paclitaxel	NCT02715804 a
		1	Lung, gastric	Pembrolizumab	NCT02563548 b
		Not applicable	Pancreatic	Cetuximab	NCT02241187 b
		Not applicable	Pancreatic	Gemcitabine + nab-paclitaxel + rivaroxaban	NCT02921022 b
		1/2	Pancreatic	mFOLFIRINOX	NCT01959139 a
		1/2	Gastric/gastroesophageal/esophageal	Leucovorin, cisplatin, paclitaxel, oxaliplatin, fluorouracil, atezolizumab, cobimetinib, ramucirumab, BL-8040, linagliptin, tiragolumab	NCT03281369 d
	Simtuzumab (LOXL2)	2	Pancreatic	Gemcitabine	NCT01472198 a
	Vismodegib (GPCR inhibitor)	2	Breast	Paclitaxel + epirubicin	NCT02694224 b
		2	Colorectal	mFOLFOX, FOLFIRI, bevacizumab	NCT00636610 a
		2	Pancreatic	Gemcitabine	NCT01195415a, NCT01064622a
		2	Ovarian, colorectal, basal cell carcinoma	FOLFIRI, FOLFOX, bevacizumab	NCT00959647 b
		2	Pancreatic	Gemcitabine, nab-paclitaxel	NCT01088815
		1	Pancreatic	Gemcitabine, erlotinib	NCT00878163 b
		2	Gastric	Oxaliplatin, leucovorin, fluorouracil	NCT00982592 a
		2	Solid tumors	Trastuzumab, pertuzumab, erlotinib, vemurafenib, cobimetinib, alectinib, atezolizumab	NCT02091141 b
		FDA approved	Basal cell	–	–
		1/2	Skin basal	Pembrolizumab	NCT02690948 a
		2	Lung	Cisplatin, cixutumumab, etoposide	NCT00887159 a
		2	Meningiomas	GSK2256098, capivasertib, abemaciclib	NCT02523014 b
	Saridegib (GPCR inhibitor)	1/2	Pancreatic	Gemcitabine	NCT01130142 a
		1	Head and neck squamous cell carcinoma	Cetuximab	NCT01255800 b
		1	Pancreatic	FOLFIRINOX	NCT01383538 b
	Pirfenidone (TGFβ inhibitor, CAFs)	1	Lung	Carboplatin + paclitaxel + pemetrexed	NCT03177291 b
		1/2	Lung	Atezolizumab	NCT04467723 d
	e.g., Losartan (Ang II inhibitor, TGFβ inhibitor, CAFs)	2	Pancreatic	Nivolumab + FOLFIRINOX	NCT03563248 d
		2	Pancreatic	FOLFIRINOX	NCT01821729 c
		2	Breast	Doxorubicin + camrelizumab	NCT05097248 d
		1	Pancreatic	Gemcitabine	NCT01276613 e
		2	Pancreatic	mFOLFIRINOX/gemcitabine + nab-paclitaxel	NCT04539808 d
		2	Pancreatic	FOLFIRINOX 1 + 9-ING-41 (NF-κB inhibitor)	NCT05077800 d
		1	Osteosarcoma	Sunitinib	NCT03900793 d
	e.g., Bosentan (endothelin I inhibitor)	1	Pancreatic	Gemcitabine + nab-paclitaxel	NCT04158635 d

^aNo change compared with standard treatment.

^bNo results posted.

^cImproved progression-free survival and overall survival.

^dRecruiting.

^eTerminated/withdrawn.

Approaches to target components of mechanical sensing and mechanotransduction

Integrin inhibitors have been investigated in Phase 1 and 2 clinical trials with promising results [62,63,100]. However, the survival benefits of chemotherapeutics combined with mechanical pathway-focused inhibitors compared with single-agent chemotherapy have been modest, with only a few exceptions improving outcomes (e.g., cilengitide, vitaxin, and dasatinib).

Targeting mechanosensors presents an alternative approach to exploit aberrant activation of mechanotransduction pathways and restore chemosensitivity. To this end, expression of the transient receptor potential (TRP) and Piezo ion channels correlates with poor patient survival, making these mechanosensors promising therapeutic targets [101,102]. In breast cancer cells, TRPM2 was linked with doxorubicin and tamoxifen resistance [103], while pharmacological and genetic inhibition of TRPC5-mediated autophagy was shown to increase the cytotoxic effect of adriamycin [104]. Furthermore, inhibitors that disrupt the ionic homeostasis of TRP channels via altering the Ca²⁺ and Na⁺ influx can induce apoptosis and inhibit cell proliferation and migration [105]. Additionally, the metabolite of ginseng saponin, 20-GPPD, stimulated Ca²⁺ influx through TRPC channel activation, which resulted in apoptosis and G1-phase arrest in colon cancer cells [106]. Cell cycle arrest has also been shown to be induced in gastric cancer cells via TRPC6 blockade and decreased Ca²⁺ influx [107]. Clinical trials against mechanosensitive ion channels are in early stages. For example, the TRPM8 activator D3263 is being tested as a single agent in a Phase 1 dose escalation study (NCT00839631). Targeting TRPV6 represents another approach to suppress Ca²⁺-mediated cancer proliferation and metastasis [108,109]. A Phase 1 clinical trial of the TRPV6 inhibitor (NCT03784677) was recently designed to investigate tolerable doses and safety of this inhibitor in patients with advanced solid tumors. A detailed review of approaches targeting TRP channels can be found elsewhere [105].

Inhibition of piezo ion channels represents another mechanosensor-targeted approach that is in early clinical development. Currently, only a few inhibitors of Piezo1 are available, with low specificity, including Gd3+, ruthenium red, and GsMTx-4 [110,111]. The G protein-coupled receptor (GPCR) family of mechanosensors could also serve as potential targets of mechanically induced chemoresistance. In mice bearing breast and lung tumors, pepducin P1pal-7, which targets the interaction of PAR1 with G proteins, was shown to block tumor progression and metastasis [112]. More recently, doxycycline was shown to selectively inhibit PAR1 in several cancer cell lines, including breast and melanoma cancer cells [113]. Furthermore, YM-254890 and FR900359 represent promising and specific inhibitors of GPCR signaling. In melanoma, YM-254890 induced tumor killing when combined with MEK inhibitors in vitro and in vivo [114]. GPCR-targeting drugs have entered clinical trial testing, such as the CXCR4 inhibitor TG-0054 [115,116].

Approaches to normalize the TME

Agents that inhibit matrix metalloproteinase (MMP)-driven ECM degradation have been tested in clinical trials to block metastatic progression. In preclinical studies, 4-methylumbelliferone (4-MU) and imatinib were shown to reduce HA accumulation in pancreatic and prostate cancers in vivo [117,118]. The subsequent reduced HA-mediated CD44 activation led to decreased PI3K, Akt, and ERK signaling, which impaired migration and invasion [117,118]. PEGylated human recombinant PH20 hyaluronidase (PEGPH20) is another therapy that has shown improved responses to gemcitabine and doxorubicin in in vivo pancreatic tumor models [119]. However, in a Phase 3 clinical trial, administration of PEGPH20 combined with nab-paclitaxel plus gemcitabine failed to improve survival outcomes [120].

Inhibiting collagen remodeling represents another ECM-targeted therapeutic strategy. Specifically, ECM remodeling involves collagen fiber crosslinking, which is mediated by the LOX family of enzymes [121]. Preclinically, inhibition of lysyl oxidases in murine breast cancer models improved doxorubicin responses by reducing tissue stiffness [122]. However, in the clinic, combination of the LOX2 antibody, simtuzumab, with gemcitabine did not improve therapeutic outcomes in patients with pancreatic cancer [123].

Targeting cancer-associated fibroblast (CAF) activation represents another strategy to alleviate CAF-mediated ECM remodeling and reduce tumor stiffness. For example, vismodegib (or saridegib) and pirfenidone, which target Hedgehog and TGFβ pathway activation in CAFs, respectively, have been shown to improve the efficacy of chemotherapeutics and immunotherapies in breast and pancreatic tumor models in vivo [124,125]. Vismodegib has already been approved by the FDA for the treatment of basal cell carcinoma; however, further investigation is needed for the treatment of other solid tumors characterized by high CAF density, including breast and pancreatic cancer. Additionally, CAF-targeted approaches involve repurposing of several FDA-approved drugs, including the antihypertensives losartan and bosentan, the corticosteroid dexamethasone, and the antihistamines tranilast and ketotifen [82,126–128].

Therapeutics that directly modulate the mechanical TME are also known as mechanotherapeutics [129]. It was shown that administering mechanotherapeutics, such as pirfenidone, losartan, tranilast, and dexamethasone, reduces tumor stiffness and mechanical forces, and increases tumor perfusion and drug delivery in preclinical tumor models of pancreatic and breast cancer [125,126,130–132]. Importantly, bosentan is being tested in a Phase 1 clinical trial in patients with pancreatic cancer (NCT04158635). Losartan has already been studied successfully in a Phase 2 clinical trial showing that, when used in combination with radiation and chemotherapy, it converted 60% of unresectable, locally advanced pancreatic ductal adenocarcinoma to a resectable status, thereby making it a ‘potentially curable’ treatment [133]. Given that only a subset of patients with PDAC respond to this therapy, further studies are needed to develop predictive biomarkers and elucidate mechanisms of treatment response.

Concluding remarks and future directions

Emerging evidence suggests that mechanical cues present in the TME affect the ability of chemotherapeutic treatments to induce cancer cell death. Recent studies have shown that mechanically induced chemoresistance is mediated through integrin or receptor tyrosine kinase (RTK) activation of prosurvival pathways, including PI3K/Akt, FAK/Src, ERK/MEK, or Wnt, which also regulate the expression of drug transporters. However, many critical questions remain (see Outstanding questions). First, the responses of cancer cells to mechanical cues may depend on the genetic alterations driving tumorigenesis, including activating mutations of the Ras, EGFR, or PI3K signaling pathways. Responses to targeted therapies against aberrant activation of these pathways may be impacted by crosstalk with mechanotransduction pathways. To this end, it would be critical to characterize directly in patient samples both cancer cells and the TME before therapy for genetic heterogeneity and mechanical alterations compared with normal tissue. Continuous monitoring of ECM stiffness, vessel function, and matrix deformation using an ultrasound imaging-based approach could provide further insights into mechanical cues acting on tumor cells [82,132]. Integrating this mechanical characterization with multi-omic profiling would enable us to dissect how TME mechanical abnormalities impact signaling responses and drug sensitivity. Importantly, a large-scale analysis of signaling pathways in surviving cancer cells and other cells present in TME, including immune cells, will lead to rational target identification for developing therapeutic approaches that target mechanoresistance mechanisms and personalized treatments.

Outstanding questions.

How do genetic alterations that drive oncogenic signaling impact transduction of mechanical cues in cancer cells?

Do mechanosensors crosstalk with different molecular complexes involved in chemotherapy mechanisms of action (e.g., disruption, transcription, or DNA replication)?

Is there overlap in the mechanically induced resistance mechanisms between chemo-, targeted, and immunotherapies?

Which mechanosensors and downstream signaling nodes should be targeted to restore sensitivity to cancer therapy when an association with mechanical cues has been established in the clinic?

Is there a therapeutic window for mechanosensitive pathways with limited on-target toxicity in normal cells compared with cancer cells?

What are the effects of mechanical cues on immune and stromal cell types in the TME with respect to promoting resistance to chemotherapy, targeted therapy, and immunotherapy?

Importantly, out of the clinical trials shown in Table 2, only a subset showed therapeutic benefits. Analysis of mechanosensor expression, such as Piezo1 or GPCRs, in multiple tissues, is critical for successful therapeutic development and strategies to enable tumor-specific delivery. Genetic studies in Piezo1-knockout mice demonstrated a lethal phenotype in utero with impaired vascular development, highlighting the importance of careful consideration when targeting mechanosensors. Similarly, targeting integrins in patients with cancer has been studied extensively, but most have not shown significant results. This could be because of the central role of integrins in normal physiology and the lack of a therapeutic window. With the emergence of more therapeutic strategies that target biophysical cues to reverse drug resistance, there is a need for experimental platforms to comprehensively explore mechanically induced drug resistance in model systems and clinically relevant samples. Such platforms could use 3D tumor cell- or patient-derived organoids embedded in a supportive matrix or exposed to mechanical forces (both compressive and tensile) within mechanical bioreactors. Cancer-on-a-chip models that allow for precise control of mechanical stimuli and fluid flow could also be used to study mechanoresistance mechanisms and test novel therapeutic strategies. Computational models using transport modeling and gene regulatory networks also present a powerful approach to study the interplay between forces, mechanotransduction, and drug responses. Finally, machine learning-based tools promise to integrate large-scale imaging and multi-omics data sets in the clinic to stratify patients based on mechanically informed signatures. Advances in these areas will lead to the identification of novel, highly targeted therapies that intercept mechanically induced drug resistance and enhance the efficacy of available treatments for patients with cancer.

Highlights.

Mechanical abnormalities present within the tumor microenvironment, including compressive forces and matrix stiffness, affect cancer cells at the molecular, cellular, and tissue level to fuel tumor growth and metastasis.

Chemotherapy, targeted therapy, and immunotherapy mechanisms of action can crosstalk with mechanotransduction signaling mechanisms that can shape drug sensitivity and resistance.

Restoring mechanical abnormalities in the tumor microenvironment and targeting mechanosensitive pathways provide opportunities for designing rational combination therapies in drug-resistant solid tumors.

Glossary

ATP-binding cassette (ABC) family multidrug efflux pumps:

family of membrane proteins that utilize energy from ATP hydrolysis to pump substances, including drugs, out of cells or into intracellular compartments

Cancer-associated fibroblasts (CAFs):

type of fibroblast present in the microenvironment of cancer tissues

Cancer stem cell (CSC):

small fraction of tumor cells that have high tumorigenic potential due to their self-renewing and tumor-initiating properties (stem cell characteristics).

Desmoplasia:

condition in which CAFs produce excessive amounts of extracellular proteins, which form a dense matrix network around and within the tumor

ECM stiffness:

material property that refers to the extent to which a material can resist deformation or deflection in response to an applied force. A relatively stiff matrix requires greater forces to deform compared with a more compliant matrix

Epithelial-to-mesenchymal transition (EMT):

biological process in which epithelial cells undergo a phenotypic change and acquire mesenchymal characteristics, including loss of cell–cell adhesion and polarity, and acquisition of migratory and invasive properties.

Extracellular matrix (ECM):

mixture of extracellular components, including collagen, fibronectin, and HA.

G protein-coupled receptor (GPCR):

large family of cell surface receptors that trigger the activation of intracellular signaling proteins, called G proteins, and initiate downstream signaling cascades

Interstitial fluid pressure (IFP):

result of increased fluid leakage from the vascular to the interstitial space due to the hyperpermeability of tumor vessels (new blood vessels formed: angiogenesis or abnormal tumor vasculature). This fluid leakage reduces blood flow and impairs drug delivery to the tumor

Mechanical forces:

mechanical stresses and pressures exerted within TME due to tumor growth and interaction of tumor cells with the surrounding tissue. In the tumor interior, these forces are compressive in all directions, while, at the tumor periphery, forces are compressive in the radial direction and tensile at the circumferential direction

Mechanotransduction:

process of translating extracellular mechanical cues into downstream intracellular signals that regulate cellular programs, cell states, and phenotypes.

Transient receptor potential (TRP):

family of transmembrane proteins that form ion channels, allowing the flow of ions across cell membranes in response to external stimuli. Upon activation, they can initiate downstream signaling cascades to regulate cellular responses

Tumor microenvironment (TME):

complex mixture of cellular and noncellular components surrounding a tumor. Cellular components include cancer, stromal, endothelial, and immune cells. Noncellular components include the ECM