The tumor microenvironment: a key player in multidrug resistance in cancer

The role of immunosuppressive components in TME

Numerous immune cells infiltrate the TME, including cytotoxic CD8⁺ T lymphocytes (CTLs), natural killer (NK) cells, dendritic cells (DCs), and various negative immune regulatory cells, including regulatory T cells (Tregs), helper T cells (Th), tumor-associated macrophages (TAMs), and myeloid-derived suppressor cells (MDSCs) [35]. The roles in tumor killing and immune suppression are shown in Figure 2.

Figure 2:

Immune cells in the TME after drug action. In the early stages of tumorigenesis, the immune system aims to destroy cancer cells, but as the tumor develops and the drugs work, fibroblasts, macrophages, T cells and B cells play the opposite role. For example, CAFs can be activated to secrete angiogenic factors, including vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF), under hypoxia and metabolic reprogramming; TAMs in the TME mostly tend to be M2-type TAMs; it can further promote tumor invasion by secreting MMP9 to support angiogenesis and invasion. TAMs are also able to inhibit Th1 immune responses by secreting CCL17, CCL18, CCL12, TGF-1, PGE2, and Il-10, protecting tumor cells from immune attack. In addition, TAMs are located around new blood vessels and can help cancer cells cross the blood vessel barrier. The figure was set up using BioRender.com.

During tumorigenesis, tumor tissue can recruit monocytes to its vicinity and transform them into TAMs [36]. In general, it is the division of activated macrophages are divided into M1 and M2 subpopulations. M1 macrophages secrete pro-inflammatory molecules such as interleukin (IL)-1, IL-12, IL-18 and IL-23, tumor necrosis factor (TNF) and chemokine C-C pattern ligand (CCL). On the other hand, M2 macrophages are primarily induced by Th2 cells and play a role in wound healing and allergic responses. They exhibit pro-tumorigenic activity and are able to downregulate the immunostimulatory factor IL-12 and upregulate the immunosuppressive factor IL-10 [37, 38]. During cancer treatment, TAMs exhibit a highly immunosuppressive M2 profile, which induces MDR through multiple mechanisms [39]. First, TAMs possess the ability to produce various cytokines that support angiogenesis, including matrix metalloproteinase (MMP) 9, MMP13, and vascular endothelial growth factor (VEGF) [40, 41]. It has been reported that mice treated with doxorubicin for breast cancer found that doxorubicin-induced tumor cell fruit TAMs secreting MMP9, which modulates vascular architecture and reduces vascular leakage by acting on substrates in TME, thereby impairing the conductive effects of doxorubicin [42]. Secondly, TAMs produce transforming growth factor (TGF)-1, prostaglandin E2 (PGE2), and IL-10, which defend tumor cells against immune attack. Production of monocyte pneumogastrin-1 (MCP-1) and interleukin 10 (IL-10) has been reported to be stimulated by TGF-1 and mediated by cross-talk between Smad, PI3K/AKT, and BRAF-MAPK signaling pathways. TGF-β1 regulates phosphorylation of Smad3 and expression of MCP-1 and IL-10 but is inhibited by P144 (a TG-β1 blocking peptide). This evidence suggests that TAMs can produce immunosuppressive effects [43]. In addition, IL-10 may activate STAT3 and enhance the expression of the B-cell lymphoma 2 (Bcl-2) gene, which can have a negative impact on tumor prognosis. TAMs may cause drug resistance in breast cancer through the IL10/STAT3/Bcl-2 signaling pathway [44]. In addition, TAMs produce prostaglandin E2, indoleamine pyrrole-2 inducer, CCL17, CCL18 and CCL22, which inhibit Th1 immune responses [45], [46], [47]. Based on micropore assays, immunocompromised mice developing melanomas co-express high levels of CCL17/22 with M2-type TAMs. Combining CD40 mAb with CSF-1R inhibitors slows tumor development, reduces the prevalence of M2-type TAMs, and increases T-cell activation by inducing the release of TNF, IL-6, and IL-12 [48]. Finally, TAMs may also release histone proteases that activate NF-κB Alternatively, they may activate IL6 and STAT3, which can lead to tumor cell-mediated resistance [49, 50].

Immature heterogeneous cells called MDSCs, which are produced in the bone marrow, may significantly impair the responsiveness of immune cells [51]. By producing nitric oxide synthase-2 (NOS-2), which reduces cysteine required for T cell activation, and arginase (ARG-1), which interferes with T cell protein synthesis, which requires arginine, MDSCs directly inhibit CD8⁺ T cells [52, 53]. Additionally, MDSCs can secrete TGF and VEGF, promoting angiogenesis and severely reducing the activity of effector T cells and NK cells, while activating T regulatory cells (Treg) [54, 55]. In tumor-bearing mice, PDE5 inhibitors have been shown to block MDSCs-mediated immunosuppressive mechanisms. As T cells activate the release of IFN-, MDSCs are receptive to IL-13 and produce their specific IFN-γ. By interacting with each other, these two cytokines activate the ARG1 and NOS-2 enzymes, leading to the release of nitric oxide (NO) and reactive oxygen species (ROS). This process generates nitrite and initiates the apoptosis of CD8⁺ T cells. PDE5 inhibitors increase the levels of cyclic guanosine monophosphate, which reduces the concentration of cell membrane Ca²⁺ concentration, leading to a decrease in calcium-dependent protein kinase C activity, which prevents upregulation of IL4Rα, which in turn impairs the IL4Rα-ARG1 pathway and allows immunosuppression to be alleviated [56]. Tumor-bearing mice with functionally suppressed MDSCs exhibit increased CD8⁺ T cell responsiveness, suggesting that these cells play an important role in tumor-induced immunosuppression [57].

Tregs, a subpopulation of CD4⁺ T cells expressing Foxp3, suppress the anti-tumor response of the immune system [58]. miR-142-3p reduced adenylate cycle 9 (AC9) expression was inhibited by Foxp3, whereas cAMP was promoted by AC9. Elevated cAMP levels inhibit antigen-presenting cells (APCs) and effector T cells through gap-functioning intercellular communication [59]. Tregs can promote immunosuppression of TME by secreting executive and proenzyme B and the immunosuppressive cytokines IL-10, IL-35 and TGF-β or by inhibiting T cell activity through direct contact [35]. By considering the effect of iNOS inhibition on Treg production in mouse and human CD4⁺ T cells, TGF1-dependent Treg activation was ameliorated in the work of Jayaraman et al. [60]. Higher numbers of Foxp3⁺ Treg were observed in tumor-bearing mice and within tumors. The combination of iNOS inhibitors and cyclophosphamide effectively inhibits MDSCs and Tregs, accelerating the metastatic process and thereby inhibiting tumor progression in an immune-dependent manner [60].

The immune system plays a crucial role in immunosurveillance as immune cells from both the adaptive and innate immune systems infiltrate the TME to regulate tumor progression. The immune regulatory cells mentioned above, including TAMs, MDSCs and Tregs, can all influence tumor progression. During drug treatment, all of these cells can respond with a corresponding regulatory response, thereby diminishing the therapeutic effect of the drug and consequently leading to the occurrence of MDR.

Soluble factors in the TME act in a paracrine manner

Following medication, damage to the TME’s DNA may trigger a reaction to stress in neoplastic and tumor-associated stomatal cells, which causes the creation of a variety of soluble substances, including GM-CSF, G-CSF, SCF, VEGF, and IL3, which promote myelopoiesis and contribute, in part, to a blockade of myeloid cell maturation [61], and amphiregulin (AREG) and epidermal growth factor (EGF) aid in epithelial cell proliferation, which also encourages the invasive WNT families, promotes tumor growth and survival, and results in increased resistance to chemotherapeutic drugs, and IL-1, IL-6, IL-8, and IL-27, which control inflammatory cells activity, and epithelial cell proliferation is aided by AREG and EGF [62, 63]. This DNA damage response stops cell growth and inhibits tumor development in a cell-autonomous, intrinsic, autocrine, or paracrine manner. Study shows that lymphoma survival following the use of a mouse model of Burkitt’s lymphoma is influenced by the paracrine component of TME. After DNA damage, IL-6 is secreted and released in a p38 MAP kinase-dependent manner, exerting a pro-survival benefit by inducing Bcl2 family members. Blocking the inhibitory effect of Bcl2, IL-6’s ability to promote doxorubicin was inhibited [64]. Additionally, it has been shown that after DNA damage, nuclear factor control of the NF-κB gene enhancer for light polypeptide activates WNT16B, which then sends a paracrine signal to tumor cells to activate the classical WNT program [65]. WNT16B expression in the prostate of TME reduces treatment efficacy and accelerates tumor growth. Cyclic chemotherapy could enhance resistance to subsequent treatment through this mechanism, contributing to the non-autonomous role of cells in TME [65].

The TME promotes metabolic reprogramming

TME is chronically hypoxic due to rapid tumor expansion, increased metabolic levels, and relatively inadequate oxygen supply, which is directly related to tumor resistance [66], [67], [68], [69]. Hypoxia-inducible factor-1 (HIF-1) binds to hypoxia-responsive elements (HRE) in the nucleus and promotes genes for a number of metabolic pathways as well as angiogenesis, growth factor, and receptor release, glycolysis, and other cellular processes [70, 71]. In colorectal cancer, it is reported that HIF-1 may improperly activate the PI3K/AKT and Wnt/β-catenin signaling pathways, leading to tumor cells’ drug resistance to 5-fluorouracil [72]. The MDR1 gene encodes P-glycoprotein, an efflux drug transport protein, and can pump hydrophobic anticancer medicines into the tumor’s outer ring while preventing drug buildup in the tumor cells, thus affecting the effective concentration of drugs in cells, resulting in the occurrence of tumor MDR [73]. Hypoxia regulates P-glycoprotein, which is a HIF-1 downstream target. P-glycoprotein expression may be decreased by HIF-1 inhibition [74, 75]. When P-glycoprotein was inhibited in hypoxia by the combination of the selective P-glycoprotein efflux activity inhibitor, tariquidar, and proteasome inhibitors, multiple myeloma cells were re-sensitized to bortezomib and carfilzomib [76]. Additionally, hypoxia particularly causes tumor cells to produce VEGF, which produces antiapoptotic proteins (including survivin and Bcl-2) and stimulates the PI3K/Akt antiapoptotic pathway [77], [78], [79]. The activation of HIF-1 and NF-κB, the promotion of CCL28 and cytokine production, and the release of VEGF are other ways that hypoxia may trigger inflammation [80]. Due to the production of VEGF, patients with lung adenocarcinoma had considerably greater blood concentrations of CCL28 than healthy individuals [80].

In addition to increased oxygen demand, there is usually an abnormal increase in glucose and glutamine requirements in tumor cells. The Warburg effect explains how tumor cells readjust their metabolism to cope with a nutrient-poor environment and support the growth of malignant tumors; even when sufficient oxygen is available, tumor cells abnormally activate glycolysis and inhibit mitochondrial respiration [81, 82]. The final product of glycolysis is lactic acid, which is transported by cells through transporters. HIF-1α can participate in the transport process by regulating monocarboxylic acid transporter 4, causing the TME to become acidic [83]. The low pH of TME (pH 6.5–6.9) promotes extracellular matrix (ECM) degradation via MMP and tissue proteases, releasing VEGF to increase angiogenesis and inhibit the tumor antigen-induced immune response, which can lead to a tumor-invasive phenotype and chemotherapy resistance [84, 85]. In addition, hypoxia and acidic TME increase the degree of dissociation of weakly basic drugs (e.g., adriamycin) and increase protonation, reducing ion trapping and drug uptake, leading to multidrug resistance [85]. Transport by the proton pump enhances the lysosome’s extracellular acidity and intracellular acidity. Under acidic conditions, the ability of the lysosome to chelate weakly alkaline drugs is enhanced and the drug is trapped in the lysosomal lumen, unable to interact with the target, leading to cellular resistance [86].

TME usually has an acidic environment, which is due to the abnormal metabolism of tumor cells, which produces a large amount of lactic acid and other acidic metabolites, leading to a decrease in pH. This acidic environment is conducive to the proliferation, invasion and metastasis of tumor cells, and at the same time can inhibit the activity of immune cells, resulting in a decrease in the ability of the immune system to attack tumor cells [87, 88]. The supply of oxygen in the TME is often uneven, with areas of hypoxia. This is due to the underdevelopment of the tumor vasculature, resulting in an inadequate supply of oxygen. Hypoxia can induce tumor cells to produce adaptive responses such as enhanced glycolysis, activation of the gluconeogenic pathway, and apoptosis and necrosis. Hypoxia can also promote malignant phenotypic transformation of tumor cells and increase their invasiveness and drug resistance. In addition, a variety of enzymes are present in TME, including protein hydrolases and oxidoreductases, which play important roles in tumor cell invasion, metastasis and immune escape. For example, protein hydrolases can break down the extracellular matrix and basement membrane, helping tumor cells to cross the tissue barrier. Oxidoreductase, on the other hand, is involved in the regulation of oxidative stress and protects tumor cells from oxidative damage [89], [90], [91]. Hypoxic and acidic niches are symbiotic and interactive, and targeting alone cannot effectively inhibit MDR. Therefore, further exploration of therapeutic strategies and specific molecular mechanisms should be part of the key research directions for TME.

The TME affects drug absorption

Drug distribution within tumors mainly occurs through diffusion, and the drug permeation barrier is one of the most significant factors that hinder drug uptake. Of the components of the TME, the ECM, a fibrous network of extracellular proteins and other macromolecules secreted by all the cells in the TME, is the most important; these components include collagen, fibronectin, proteoglycans, elastin, and integrins [92]. In normal tissue, the ECM is loose and consists of collagen, elastin, and glycoproteins, which can provide stable tissue structure and the circumstances required for cell division and proliferation. In most solid tumors, however, the ECM consists of another form of connective tissue fiber that becomes denser and stiffer. In addition, TME increases collagen synthesis, promotes cross-linking with elastin, and enhances the deposition of ECM, resulting in increased tumor stiffness [93], and proteoglycans combine with collagen to form cartilage components, both of which contribute to the stiffness and firmness of the tumor tissue [94]. It overabundantly surrounds tumor cells with this stiff, thick ECM, which serves as a barrier to keep medications out of the cells. However, since blood and lymphatic arteries are often absent from thick ECM, the transport of oxygen, nutrients, and metabolites is hampered. Pathways by which hypoxia and increased metabolic load trigger antiapoptotic and drug resistance mechanisms [95]. The production of crosslinks between collagen molecules and elastin networks may be facilitated by five families of lysine oxidase isozymes (LOX and LOXL1-4). Therefore, an increase in lysine oxidase can affect drug distribution [96, 97]. It is possible to simulate hypoxia and the tumor microenvironment in a way that uses a 3D in vitro cellular model in order to study tumor immune escape mechanisms and demonstrate its impressive potential in drug discovery. The model successfully simulated the survival and adaptability of tumor cells under hypoxic conditions and their interactions with the immune system [98]. Inhibition of lysine oxidase can improve oxygenation, as discovered in research on colorectal, ovarian, and breast cancer [96]. By oxidizing the extracellular domain of the receptor for platelet-derived growth factor, several investigations have shown that lysine oxidase may directly influence the production of vascular endothelial growth factor A (VEGF-A) [99].

TAFs are one of the most critical components of TME and are essential for altering ECM. There are several sources of TAFs, including local fibroblasts, stellate cells, bone marrow mesenchymal stem cells, and cells derived from adipose tissue [100]. Although TAFs exhibit significant heterogeneity, they share the same biological functions: invasiveness, proliferation, and promotion of immune responses. TAFs can promote MDR through a variety of mechanisms [94, 101, 102]. This section mainly describes how TAFs interfere with tumor therapy by participating in ECM remodeling.

The production of a range of matrix metalloproteinases (MMPs), including MMP-1 and MMP-3, as well as the secretion of several matrix proteins, including fibronectin and type I collagen, by TAFs may contribute to degrade the normal ECM structure and stiffen the matrix. TAFs can secrete proteoglycan growth factors, promote ECM remodeling, participate in the formation of therapeutic barriers through interactions with integrin and cadherin, and hinder antineoplastic drugs and immune cells in clinical treatment [103, 104]. Hyaluronic acid (HA) causes interstitial fluid pressure to be high which contributes to extracellular matrix remodeling, reducing drug exchange through capillaries [105]. In a TGF-dependent manner, TAFs can facilitate HA synthesis by promoting EGF production, which then triggers the ERK-MAPK pathway linked to EGFR [106], [107], [108]. According to some studies, TAFs inhibit the ability of cancer cells to respond to chemotherapy by activating interleukin-1 receptor-associated kinase 4 (IRAK4), promoting NF-κB activity in the stroma, and additionally can affect the transport of the gemcitabine drug in a mouse model of pancreatic cancer [109], [110], [111]. In addition, pro-epidermal growth factor, hepatocyte growth factor (HGF), and stromal cell-derived factor-1 (Sdf-1) are just a few of the cytokines that TAFs may release, which activate a series of cascade reactions and eventually lead to MDR and tumor recurrence [112]. In lung cancer, TAFs secrete IL-6, IL-8, and HGF and activate the miR-21 and MEK/ERK pathways, which promotes resistance to EGFR tyrosine kinase inhibitors [113]. Sphingosine-1-phosphate (S1P) and prostaglandin E2, which result from COX-2 activation and are secreted by TAFs as well, operate either autocrine or paracrine to promote apoptosis and chemical resistance via PI3K-Akt/PKB pathway activation [114].

Another important factor affecting drug absorption is angiogenesis around the tumor. The design and purpose of the newly formed blood arteries at the tumor location were found to be abnormal, and the blood flow was blocked substantially [115]. The tumor vasculature is also highly permeable and leaky, favoring non-specific extravasation of blood components into the surrounding interstitial space. As a result, the TME is not only hypoxic and acidic but also surrounded by high interstitial pressure, which acts as a pathological barrier, preventing drugs from entering the tumor and leading to MDR [116]. On the other hand, leaking blood vessels can promote the extravasation of the drug in the tumor area, resulting in an increased concentration of the drug in the tumor, leading to an enhanced therapeutic effect, but accelerating the formation of MDR. Drug distribution is also affected by blood vessels’ aberrant shape and function. Due to the obstruction of blood flow, the drug concentration in contact with tumor cells makes it difficult to produce cytotoxicity [117]. And the insufficient supply of nutrients leads to difficulties in excreting metabolites, affecting drug efficacy [117]. In addition to the secretion of VEGF, vascular endothelial cells may also have an impact on the expression of adhesion molecules on immune cells and endothelial cells, such as the integrin ligand intercellular adhesion molecule 1 (ICAM 1) and vascular cell adhesion protein 1 (VCAM 1). An immune checkpoint protein called programmed cell death 1 ligand 1 (PD-L1) binds to PD-1 expressed on T cells, inhibiting its anti-cancer activity, possibly produced by tumor-associated production, and influencing resistance to tumor therapy by inhibiting anti-tumor capacity [118]. In Zhang et al.’s study [119], the effects of VEGF overexpression and VEGF receptor 2 inhibition on the response to doxorubicin in soft-tissue sarcoma were studied in immunodeficient mice. The outcomes demonstrated a high degree of vascular proliferation, a shorter incubation time, quicker development, greater chemical resistance, and a higher incidence of lung metastasis in the xenografts with high VEGF expression. Treatment with the anti-VEGFR2 monoclonal antibody resulted in a reduction in the number of microvessels and an increase in the vascular maturation index, which also facilitated the response to doxorubicin and inhibited VEGFR2 signaling. The above evidence suggests that chemoresistance in soft-tissue sarcomas is caused by the interaction between tumor-associated endothelium cells. on the other hand, antiangiogenic therapies such as bevacizumab can lead to a further decrease in the vasculature, conferring hypoxia and physical seclusion from circulating drugs [120].

Angiogenesis, the creation of formation new blood vessels from pre-existing ones, plays an important role in tumorigenesis. Benign tumor cells exist in a dormant state and are affected when they have difficulty obtaining an adequate blood supply. However, an ‘angiogenic switch’ occurs when dormant tumor cell angiogenesis is activated, and factors are secreted that induce endothelial cells to germinate and converge into tumor masses. In the hypoxic environment of the internal tumor mass, the dimeric protein complex of HIF-1 remains stable and activates the expression of numerous genes that contribute to the angiogenic process, including vascular VEGF, which promotes vascular permeability, and bFGF, which promotes endothelial cell growth [121, 122].

Thus, the microenvironment itself is not always an active ‘soil’ per se, but occurs only incidentally in the vicinity of tumor multiplication, and tumor angiogenesis is a complex process of pro- and anti-angiogenic signaling pathways generated by both malignant and non-malignant cells through autocrine and paracrine production. Vascular density can have both positive and negative effects on drug delivery and efficacy.

Interaction between the TME and cancer stem cells

A limited subpopulation of cancer cells known as CSCs has an extensive capacity for self-renewal and division, which may contribute to the development and spread of malignant tumors [123]. They resemble embryonic stem cells [123]. In colorectal CSCs, HIF-1 has been shown to work in conjunction with the TAFs paracrine signal TGF-2 to activate the hedgehog transcription factor GLI2. This enhances dry/de-differentiation and genetic resistance to chemotherapy [124]. CSCs themselves are insensitive to conventional treatments such as radiotherapy and chemotherapy, showing drug resistance characteristics. In addition, MDR cells may develop resistance to cytotoxic compounds to which they have never been exposed. MDR may involve a variety of biochemical mechanisms [125, 126]; the most common MDR phenotype, also referred to as classic MDR, is associated with overexpression of transmembrane proteins that membrane proteins extrude substrates in a unidirectional manner, reducing the intracellular concentration of cytotoxic drugs below the effective dose. In MDR tumor cells, most of the overexpressed transporter proteins belong to the ATP-binding cassette (ABC) family of transporters, which actively transport substrates, including chemotherapeutic drugs, against a concentration gradient using energy generated by ATP hydrolysis [127]. Indeed, overexpression of these proteins is one of the hallmarks of drug resistance.

ABC transporters, which may move chemotherapy medications out of cells, lower intracellular drug concentrations, and lessen the risk of drug-induced cell toxicity, are extensively expressed by CSCs [128, 129]. Chemotherapeutic drugs primarily act on DNA or its replication process, destroying proliferating tumor cells by blocking the production of DNA and RNA, adding alkyl groups, or inhibiting enzymes. However, most CSCs are in the G0 phase and do not produce DNA or undergo cell division, which allows them to escape the killing effect of chemotherapeutic drugs and survive [130]. The damage repair ability of CSCs is robust. Even after an injury, p53, ATM-Chk2, ATR-Chk1 DNA, and other damage response pathways can be activated, leading to cell cycle arrest in G1, S, and G2 phases. This allows sufficient time and opportunity for the cells to undergo repairs [131]. The maintenance of the stem-like characteristics of CSCs is the main cause of drug resistance.

As a result of sustained external stimulation, the TME is altered and CSCs are enriched, leading to more resistant tumor tissue. The TME has a significant hypoxic component, which is essential to maintain CSC stemness. In a hypoxic environment, numerous signaling pathways mediated by HIF-1 are activated and involved in various aspects of cellular metabolism, leading to drug resistance. In human glioma cells, HIF-1 α is upregulated in CD133-positive CSCs, promoting their ability to self-renew. Hypoxia-mediated growth of CD144-positive CSCs is prevented after HIF-1 knockdown. Hypoxia-driven CD133 amplification was decreased when the PI3K-AKT or ERK1/2 pathways were inhibited, indicating that these signaling pathways may regulate hypoxia responses. Finally, CSCs cultured under hypoxia preserved an undifferentiated phenotype, unlike those cultured under normoxia. These findings imply that CSCs respond to hypoxia by activating HIF-1 to enhance the self-renewal capacity of CD133-positive cells and prevent the differentiation of CSCs [132]. Hypoxia also maintains the quiescent state of CSCs, inhibits apoptosis, and makes cells resistant to the killing effect of the drug. In non-small cell lung cancer (NSCLC), the CSC population can be reduced by modulating the PI3K/AKT pathway to increase the anticancer effect of gefitinib and reverse drug resistance [133]. In addition, TAFs and M2 macrophages in TME can secrete different cytokines and growth factors, including VEGF, SDF-1, IL10, and IL7, which can regulate the proliferation of CSCs and maintain their populations. Vermeulen et al. [134] found that HGF secreted by TAFs activates the Wnt signaling pathway through its interaction with c-Met. This activation leads to the development of stem-like properties in tumor cells located at the periphery of colon cancer. Consequently, these cells become resistant to medication and contribute to the recurrence of colon cancer. In breast cancer, continuous NF-κB activation and prolonged P65 nuclear retention are required to maintain CD10GPR77 TAFs’ paracrine IL6/IL8 secretion function and to provide a survival niche for CSCs; to provide a suitable survival microenvironment for CSCs, promote the development of tumor and drug resistance [135]. In addition, tumor cells can also secrete IL-6, which promotes the secretion of large amounts of CXCL7 by mesenchymal stem cells, and CXCL7 can promote the secretion of IL-6 by tumor cells, thus creating a cycle; in this process, both tumor and mesenchymal stem cells produce many cytokines, including CXCL1, CXCL5, CXCL6, IL-8, and IL-6, which promote the self-replication capacity of tumor stem cells [136].

Targeting the immune system of the TME

Normally, T cells use the tumor’s immune circuit to recognize and clear tumor cells from the body, but tumors use immune checkpoints, regulatory molecules that suppress the immune system, to create an immune escape [137]. Immune checkpoint blockade (ICB) aids immune cells in their anti-tumor actions by designing relevant antibodies to inhibit immune escape. It has been documented that one of how immune escape occurs is through the interaction of the programmed apoptotic protein PD-1 with the ligands PD-L1 or PD-L2, thereby preventing T-cell activation [138]. Thus, in TME, the binding of PD-1 to ligands may be reduced by overexpression of PD-L1 or PD-L2, or by PD-1 inhibitors (e.g., pembrolizumab), thereby suppressing the intratumoral immune response and promoting T-cell activation [139]. In addition, anthracyclines, cyclophosphamide, oxaliplatin, and statins can induce tumor immunogenic cell death (ICD), release DAMPs and stimulate immune responses via APC effector T cells (e.g., DCs) [140]. To date, numerous studies have shown that the combination of chemotherapy and ICB is more effective. For example, pembrolizumab in combination with chemotherapy (NCT02039674) is the first-line therapy for metastatic NSCLC, with response rates rising by nearly half compared to monotherapy and the likelihood of disease progression reduced by nearly half [141]. Pembrolizumab in combination with pemetrexed and platinum is the first-line therapy for tumors lacking EGFR and ALK gene mutations in patients with metastatic NSCLC [142]. Pembrolizumab in combination with the conventional chemotherapeutic agents carboplatin, paclitaxel and paclitaxel, whose approval was based on the results of the phase III clinical study KEYNOTE-355 (NCT02819518), for the treatment of incurable, locally recurrent or spread triple-negative breast cancer, reducing the risk of disease progression and death by 35 % [143].

In addition, studies have shown that the combination of VEGF and ICB is effective because immune cells are unable to penetrate the tumor due to their vascularity and ECM. Atezolizumab, a PD-L1 monoclonal antibody, in combination with bevacizumab and chemotherapy can be effective in treating metastatic non-squamous NSCLC, but in patients without EGFR or ALK mutations [144]. Meanwhile, bevacizumab combined with TH-302, a drug activated in hypoxia, can be used for clinical effect in treatment-resistant glioblastoma multiforme, according to early findings from a phase II clinical study (NCT02342379) [145]. First-line treatments for advanced renal cell carcinoma (RCC) including tyrosine kinase inhibitors or cytokine therapy have failed. Fortunately, according to the results of the phase III KEYNOTE-426 study (NCT02853331), pembrolizumab combined with a tyrosine kinase inhibitor (axitinib) could be used as first-line treatment in advanced RCC, superior to sunitinib, and its combination saved almost half of the patients who died [146]. Meanwhile, results from the phase III JAVELIN Renal 100 trial (NCT02684006) suggest that the PD-L1 monoclonal antibody album in combination with axitinib could be used as a first course of treatment for patients with advanced RCC [147]. Further, pembrolizumab in combination with lenvatinib treatment was effective in controlling renal, endometrial and hepatocellular carcinoma, with control rates of 83–100 % for these three diseases [148]. The study found that the use of a new drug targeting the DNA repair pathway AZD6738, an ATR inhibitor [149]. And combining tazemetostat with other treatments may work better in treating patients with hematological malignancies and may improve disease response and durability of response. In addition, the study is to assess the safety and tolerability, obtain maximum tolerated dose (MTD) and/or the recommended phase 2 dose (RP2D) of LM-101 as a single agent or in combination in patients with advanced malignant tumors.

In addition, another way in which immune escape can occur is that tumors can establish an immunosuppressive environment by producing ligands for immune checkpoint receptors, thereby limiting T cell activation [150], such as by relying on young plasmacytoid dendritic cells (pDCs) [151]. The literature suggests that the role of Tregs and MDSCs with immunosuppressive potential is supported by immature pDCs, which lack type I interferon expression [152]. In response to this phenomenon, the TLR9 agonist CMP-001 was developed to activate immature pDCs by binding to the TLR9 receptor, enabling them to produce large amounts of type I interferon while promoting the production of other co-stimulants and presenting to T-cell tumor antigens, producing an efficient anti-tumor T-cell response [153]. Phase Ib study (NCT02680184) Results showed that the combination of CMP-001 and pembrolizumab for the treatment of patients with metastatic melanoma was significantly more effective than historical data on the efficacy of pembrolizumab alone, and a phase II clinical trial (NCT04698187) of the combination is therefore underway [154]. A phase I clinical study (NCT01885897) in which 19 % of patients with hematological malignancies had a clinical response to ALT-803 drug therapy with good treatment outcomes [155]. Phase Ib dose-escalation clinical study (NCT02523469) to assess the effectiveness of combination therapy with ALT-803 in combination with nivolumab in patients with NSCLC and safety and is currently in a phase II trial [156]. NKTR-214 and nivolumab are being used as first-line therapy in metastatic melanoma in PIVOT-02, a dose-escalation study (NCT03635983) that is investigating the safety and immunological efficacy of both agents with good tolerability, independent of baseline PD-L1 status [157].

Tumor-infiltrating lymphocyte (TIL) therapy has recently attracted a lot of interest due to the development of CAR-T therapy. TIL are TIL cells isolated from tissue near a tumor, expanded in vitro by the addition of the growth factor IL-2, and then infused back into the patient to amplify the immune response and treat primary or secondary tumors. TIL can kill primary or secondary tumors directly by releasing cytotoxic tumor cells. In addition, it can control the immune response and enhance the body’s ability to eliminate cancer cells [158]. In a multicentre phase II clinical trial, it was found that in patients with advanced cervical cancer, those who received TIL in addition to chemotherapy were more likely to respond to treatment [159]. Data from the study showed a response rate of 1 in 3 and a disease control rate of more than half for ICB treatment with innovative-01 in melanoma patients who had failed previous treatment [160]. This result suggests that TIL has the potential to overcome resistance to ICB.

Targeting other components of the TME

To improve the overall uptake and uniform distribution of drugs in tumor cells, researchers have developed drug delivery systems (DDS) for systemic and/or local drug therapy [161]. Assembled from carriers, antitumor drugs and drug efflux pump inhibitors, DDSs can achieve better therapeutic outcomes by cooperatively addressing different biological signaling pathways, allowing patients to receive low doses of chemotherapeutic drugs in combination. In addition, multiple drugs delivered together can target the same cellular pathway to maximize efficacy. Nanoparticle (NP) carriers prevent drug molecules from being recognized by efflux pumps, and intracellular drug release minimizes the rate of efflux-based drug transport in vitro [160]. Physical properties such as ECM and ECS in the tumor microenvironment can affect the diffusion and penetration ability of nanoparticles. The composition and structure of ECM and ECS in tumor tissues are more complex, including collagen, elastin, fibronectin, etc. Abnormal accumulation and distribution of these components may hinder the diffusion and penetration of nanoparticles. Secondly, chemical factors such as acidity, redox state and cytokines in the tumor microenvironment can also affect the drug delivery efficiency of nanoparticles [162], [163], [164], [165], [166], [167], [168]. For example, the acidic environment in the tumor tissue may affect the stability and drug-release behavior of nanoparticles, while cytokines may influence the interaction and uptake efficiency of nanoparticles with tumor cells. In addition, biological factors such as immune cells and stromal cells in the tumor microenvironment may also have an impact on the drug delivery of nanoparticles. For example, immune cells such as TAMs and DCs may affect the phagocytosis and processing of nanoparticles, which in turn may affect drug efficacy [169], [170], [171], [172], [173], [174]. Therefore, in recent years, researchers have increasingly investigated nanodrug delivery systems (NDDSs) [175]. NDDSs involve the direct delivery of drugs to the tumor, thereby reducing toxicity and bypassing many NDDSs including Doxil [176] and Abraxane [177]. However, hypoxia, low pH, and intercellular pressure in TME are important causes of MDR and can also lead to suboptimal therapeutic efficacy of foreign vector NDDSs. Therefore, there is a need to improve the oxygenation of the tumor. A study assessed the ability of CRLX101, an investigational nanoparticle-drug conjugate containing the payload camptothecin (CPT), to improve therapeutic responses as compared to standard chemotherapy [178]. EP0057 consists of a sugar molecule cyclodextrin linked to a chemotherapy drug called camptothecin. The combined molecule or “nanoparticle drug conjugate” travels through the blood. Once inside cancer cells, the chemotherapy drug is released from the molecule. Olaparib is a drug that may stop cancer cells from repairing the DNA damage caused by chemotherapy. Researchers describe the safety of EP0057 and olaparib, as well as evaluating the combination for the treatment of specific types of lung cancer in small cell lung cancer (SCLC) [179].

In addition, tumor-derived exosomes play an important role in the cancer immune microenvironment and cancer immunotherapy [180]. In terms of the immune microenvironment, tumor-derived exosomes can influence the function and status of a wide range of immune cells. For example, they can activate endothelial cells to support tumor angiogenesis and thrombosis, convert fibroblasts and bone marrow mesenchymal stem cells into myofibroblasts, and thereby promote tumor angiogenesis and metastasis. In addition, tumor-derived exosomes can create an immunosuppressive microenvironment by inducing apoptosis, impairing the function of effector T cells and NK cells, inhibiting DCs differentiation, expanding MDSCs and promoting Treg cell activity [181], [182], [183], [184]. In immunotherapy, exosomes can be used as targets for cancer immunotherapy. Several studies have identified that exosomes of tumor origin can carry tumor-associated molecular and genetic information that affects other cells and tissues in the tumor microenvironment, thereby influencing tumor development and metastasis. Therefore, the efficacy of existing immunotherapeutic strategies can be enhanced by interfering with the secretion and function of exosomes. For example, inhibiting exosome production can reduce the immune escape of tumor cells and improve the killing effect of immune cells; interfering with the function of exosomes can reduce the drug resistance of tumor cells and improve the efficacy of chemotherapeutic drugs [185], [186], [187].

Reconstruction of the vascular system and reprogramming of the ECM can be effective in alleviating the abnormal stress of TME [11, 188], [189], [190], [191], [192]. The approval of the anti-angiogenic drug bevacizumab for the treatment of metastatic colorectal cancer indicates confidence in this treatment modality [193]. In clinical practice, anti-angiogenic drugs are often used in combination with tyrosine kinase inhibitors to prevent MDR [164]. In addition, the combination of anti-angiogenic drugs and immune checkpoint inhibitors can partially alleviate immunosuppression and improve therapeutic efficiency [194]. Anti-angiogenic drugs can also increase the delivery of NDDS by inducing vascular normalization [195], [196], [197]. In addition to vascular systemic remodeling, ECM remodeling plays an important role in metabolic remodeling. Losartan, an angiotensin receptor inhibitor, has been shown to inhibit TGF-β signaling, target collagen and hyaluronic acid synthesis in the ECM, and slow fibrosis [198], [199], [200], [201]. A phase II clinical trial (NCT01821729) showed that the neoadjuvant chemotherapy regimen FOLFIRINOX combined with losartan reduced advanced pancreatic ductal carcinoma [202]. Other agents with anti-fibrotic properties, such as dexamethasone [203] and metformin, have also been used to restore ECM in TME [204, 205].

TAFs-based stromal cells and neovascular systems promote tumor progression and drug resistance. Therefore, reducing the impact of TAFs and reversing the distortion of the neovascular network is essential to overcome TME component-induced MDR. Cibucizumab, a fibroblast activation protein-α(FAP-α) specific antigen that targets the surface of TAFs, has been shown to maintain disease stability in patients with FAP α-positive tumors in phase I/II clinical studies, but the efficacy of monotherapy remains to be determined [206]. FAP and TAF DNA vaccines increased infiltration of CD8⁺ and CD4⁺ T cells into tumors and, in addition, targeting protein tyrosine kinase-2, which is involved in stromal fibrosis in mice, increased effector T cell infiltration and drug delivery, thereby enhancing immunotherapy and prolonging survival in mice [207]. Simlukafusp alfa (FAP-IL2v. RO6874281/RG7461) is an immunocytokine-containing antibody against FAP and a variant of IL2 that predisposes to IL2R βγ. The triple combination of FAP-IL2v with an anti-PD-L1 antibody and an agonistic CD40 antibody was most efficacious. These data indicate that FAP-IL2v is a potent immunocytokine that potentiates the efficacy of different T-cell and NK-cell-based cancer immunotherapies (NCT02627274 NCT03875079) [208].

Regardless of the method used, targeting TME may have a more pronounced and definitive modulatory effect on the MDR response than directly targeting tumor cells. The development and composition of TME are complex and variable, while the response and adaptation to various therapies remain unclear, and many non-tumor host cells with broad-spectrum functions are also present in TME. Our understanding of TME remains limited, which poses a challenge to elucidate the mechanisms of TME-mediated MDR, identify specific target cells and develop subsequent interventions.