Multi-stage mechanisms of tumor metastasis and therapeutic strategies

Abstract

The cascade of metastasis in tumor cells, exhibiting organ-specific tendencies, may occur at numerous phases of the disease and progress under intense evolutionary pressures. Organ-specific metastasis relies on the formation of pre-metastatic niche (PMN), with diverse cell types and complex cell interactions contributing to this concept, adding a new dimension to the traditional metastasis cascade. Prior to metastatic dissemination, as orchestrators of PMN formation, primary tumor-derived extracellular vesicles prepare a fertile microenvironment for the settlement and colonization of circulating tumor cells at distant secondary sites, significantly impacting cancer progression and outcomes. Obviously, solely intervening in cancer metastatic sites passively after macrometastasis is often insufficient. Early prediction of metastasis and holistic, macro-level control represent the future directions in cancer therapy. This review emphasizes the dynamic and intricate systematic alterations that occur as cancer progresses, illustrates the immunological landscape of organ-specific PMN creation, and deepens understanding of treatment modalities pertinent to metastasis, thereby identifying some prognostic and predictive biomarkers favorable to early predict the occurrence of metastasis and design appropriate treatment combinations.

Fig. 1

Overview of the metastatic cascade in bone metastasis. In patients with tumors, a large number of cancer cells are released in the circulation on a daily basis. The process of a portion of cancer cells moving away from the primary tumor site to form a secondary tumor at a secondary site is called “metastasis”. Cancer metastasis mainly includes five steps: invasion, intravasation, circulation, extravasation, and colonization. CTCs break through the basement membrane matrix enclosing the cancer nest, intravaste the surrounding blood vessels, and circulate in the blood, where they endure physical pressure and are attacked by immune cells. Upon reaching secondary sites, CTCs undergo extravasation and infiltrate secondary locations. Successful extravasation depends on interactions between cancer cells and endothelial cells. Subsequently, cancer cells adapt, and proliferate in the metastatic organ, gradually forming metastatic lesions. The ability of cancer cells to invade adjacent tissues and establish distant colonies is a hallmark of malignancy. Advances in understanding the various stages of tumor metastasis have revealed key molecular mechanisms, such as changes in cell adhesion molecules, EMT, and interactions with ECM components and immune cells within the microenvironment. Comprehensive knowledge of the multi-stage process of tumor metastasis from invasion to colonization is beneficial for identifying novel therapeutic targets and interventions aimed at disrupting metastatic progression and improving patient survival rates

Fig. 2

The research journey of multi-stage metastasis: a timeline perspective. Metastasis, the deadliest hallmark of cancer, stands as one of the pivotal questions of the 21st century, demanding precise elucidation of its molecular underpinnings. Researchers have made significant strides in elucidating fundamental concepts of multistage tumor metastasis, molecular markers, and cellular interactions driving metastatic dissemination. From early observations of tumor spread in the 19th century, such as Stephen Paget’s seminal “seed and soil” hypothesis, to today’s utilization of advanced imaging modalities and single-cell sequencing technologies, each milestone reflects our progressing understanding of the metastatic process. Clarification of key signaling pathways such as EMT, angiogenesis, and immune evasion mechanisms has provided crucial insights into how cancer cells acquire migratory and invasive capabilities. Ongoing research efforts, including investigations into the role of the tumor microenvironment, EV-mediated intercellular communication, and the impact of genetic heterogeneity on multistage metastasis, will continue to unveil new aspects of organ-specific metastasis. In this figure, we focus on key discoveries and milestones in cancer metastasis research, illustrating the timeline of research history and highlighting trajectories of discovery

Fig. 3

The metabolic interventions with immune cells exited in TME. Due to the limited availability of oxygen, nutrients, and other substances in the TME, it presents a challenging milieu where cancer cells must adapt to survive under harsh conditions. In response, these cells undergo metabolic alterations involving three main nutrients: carbohydrates, amino acids, and lipids. For instance, genes involved in cellular fatty acid uptake (CAV1, CD36) and de novo synthesis (PPARA, PPARD, MLXIPL) are frequently amplified specifically in metastatic tumors. Lipids synthesized de novo can modulate membrane fluidity, impacting interactions between tumor cells and immune cells, thereby exerting anti-tumor phagocytic functions. Additionally, these lipids can act as signaling molecules, triggering oncogenic cascades. Deprivation of Gln in the TME is known to impair differentiation of Th1 cells, a subset of T helper cells crucial for coordinating anti-tumor immune responses. Moreover, enzymes such as indoleamine 2,3-dioxygenase 1 (IDO1) catalyze tryptophan oxidation, inducing T cells into the G1 phase of the cell cycle and fostering Fas-mediated cell apoptosis. In the prevalent hypoxic conditions of solid tumors, oxygen deprivation coordinates TCR stimulation and mitochondrial dysfunction in T cells, resulting in a state of exhausted T cells that suppress anti-tumor immunity. Furthermore, acidic pH levels in the TME exacerbate immune suppression by altering the metabolic pathways of immune cells. Low pH induces metabolic dysregulation in T cells, activating checkpoint molecules and promoting immune suppression. Additionally, low pH inhibits mTOR and NK cell anti-tumor activity, suppressing expression of iNOS, CCL2, and IL-6 in M1-type macrophages. In Treg, the transcription factor Foxp3 is conducive to the oxidation of L-lactic acid to pyruvate. Meanwhile, the accumulated lactic acid can enhance oxidative phosphorylation and the oxidation of nicotinamide adenine dinucleotide by inhibiting Myc and glycolysis, thus participating in the metabolic reprogramming process of Treg cells. Furthermore, activation of Toll-like receptor (TLR) signaling in tumor cells disrupts cAMP production, enhancing anti-tumor immune responses. Understanding the metabolic complexity within the TME is crucial for developing effective therapeutic approaches. Targeting these aberrant metabolic reprogramming processes holds promise for enhancing current immunotherapies and improving outcomes for cancer patients

Fig. 4

Pro-tumor functions exerted by various immune cells during tumor progression. a The remodeling process of the tumor ECM is regulated by macrophages. WNT/β-catenin signaling regulates the release of inflammatory factors. Meanwhile, IL-6 can induce EMT to enhance CRC migration and invasion. In HCC, IL-8 stimulates M2-type polarization of TAMs, promoting EMT. The migrating tumor cells were preferentially located near the tumor microenvironment of metastasis (TMEM) gate after escaping from the tumor cell nest. b Neutrophils transport lipids into tumor cells through the macropinocytosis lysosome pathway. Th2 cell-derived IL-4/IL-13 promotes the formation of NETs to reshape PM niches. Cathepsin C (CTSC) promotes the formation of NETs. HCC-induced NETs activate TLR4/9 while inducing an inflammatory response by up-regulating COX2. Additionally, NETs can bind to CCDC25 on cancer cells as a chemokine. IL1β and IL6, as well as Vcam1 gene transcripts, play an important role in the formation of CTC-neutrophil clusters. TLE1 mutations in CTCs increase G-CSF and form a positive feedback loop with other cytokines. c NK cells exist to counteract the mechanism by which tumor cells down-regulate the expression of MHC I. The function of NK cells is regulated by both active and inhibitory receptors. Silencing of NKG2DL can lead to the failure of NK cells to activate, inducing potential immune evasion in SCLC and neuroblastoma. Tumor-derived molecules, tumor-associated stromal cells, and tumor cells exert inhibitory effects on NK cells. IFN-γ production and the overall amount of IFN-γ positive NK cells in the lungs were substantially reduced in mice treated with IL-33 which could diminish NK cells’ capacity when combined with type 2 innate lymphoid cells (ILC2). d TGF-β induces cancer cells to produce IL-17RB. Knockout of Blimp1 in Treg reprograms it into responsive T cells, promoting IgE deposition and secondary macrophage activation process. Clearance of Tregs restores the function of CD8+ T cells based on the significantly increased expression of ICOS, IFNγ and CD107a. CircUSP7 can inhibit the secretion function of CD8+ T cells or the expression of Src homology region 2-containing protein tyrosine phosphatase 2 (SH2P2). Matrix Gla protein (MGP) enriches intracellular free calcium and promotes CD8+ T cell depletion. IgG activates the NF-κB pathway and promotes tumor metastasis. Down-regulation of CXCR4 and VLA4 leads to premature emigration from BM. Thymic stromal lymphopoietin (TSLP) induces B-cell precursors to differentiate into Breg, while M-CSF promotes their differentiation into macrophage-like cells (B-MF). IL21-secreting Tregs stimulate B cell activation, and the granzyme B produced can degrade part of TCR

Fig. 5

Influences on tumor invasiveness from lifestyle, neurological, environmental, aging, and circadian perspectives. The invasiveness of tumors is influenced by multiple factors, including genetic heterogeneity, tumor immune microenvironment subversion, and systemic macroenvironmental modulation. In addition to recognized dietary risk factors, lifestyle factors such as smoking, alcohol consumption, exercise, and sleep play significant roles in the proliferation, invasion, and progression of tumors. For instance, peripheral neurons within the TME can secrete neuropeptides that activate normal aHSCs, thereby promoting invasion and metastasis of HCC. Specific MANF deficiency in the liver upregulates Snail1 and Snail2 levels, thereby promoting EMT and accelerating HCC progression. In the context of PDAC, the invasion of DRG cells depends on the expression of ANXA2 and axon guidance molecule SEMA3D. Functional modulation of ANXA2 influences SEMA3D secretion and enhances tumor cell migration and invasion by binding to PLXND1 receptors on DRG surfaces. Furthermore, long-term exposure to incomplete combustion products such as ultrafine particles of carbon black in air increases glycolysis and lactate production, resulting in an immunosuppressive microenvironment. Experimental observations reveal that bacterial communities predominantly inhabit microecological niches with lower vascularization and higher immunosuppression. Additionally, cell populations lacking HPV expression exhibit reduced HPV-related cell cycle phenotypes, weaker treatment responses, and enhanced invasive capabilities. Bacterial metabolite TMAO activates the PERK pathway to induce ferroptosis in tumor cells, thereby enhancing CD8+ T cell-mediated anti-tumor immune responses. With advancing age, invasive cancer cells produce increased levels of MMA, inducing SOX4-related remodeling of the TME, activating fibroblasts, and reciprocal secretion of IL-6-carrying EVs involved in cancer progression. In human tumor cells undergoing therapy-induced senescence, upregulation of IFN-γ receptors triggers CD8+ T cell-mediated tumor rejection, enhancing the efficacy of immunotherapy. Disruption of circadian rhythms plays a crucial role in T cell exhaustion, with malignant cells exhibiting enhanced glycolysis and EMT activation linked to high circadian disruption scores. Deleting key clock transcription factor BMAL1 exacerbates fibrotic phenotypes across various tumors. Finally, in glioblastoma stem cells, strict regulation of self-renewal by the BMAL1 gene supports optimal cell growth. Resistance to AR-targeted therapy in PCa cells correlates with extensive reprogramming of FOXA1 loci and enrichment of clock component ARNTL. Rhythmic transport of DC to tumor-draining lymph nodes controls CD80-dependent circadian responses of specific CD8+ T cells, thereby enhancing therapeutic outcomes

Fig. 6

Perspectives from immunology on the organ-specificity of PMN formation. a Chronic stress activates pulmonary epithelial cells to secrete ACh, promoting lung NETs production; extracellular vesicle RNA of BC can upregulate TLR3 expression in alveolar epithelial cells, thereby triggering NETs formation. Elevated TLR3 pathway expression is also associated with smoke-induced chronic inflammation. Air pollutants induce autophagy-dependent TRIM37 degradation in alveolar epithelial cells, promoting neutrophil recruitment. Reduced let-7s in Lin28B EVs also participate in regulating neutrophil recruitment. Furthermore, low let-7s in EVs and nicotine promote lung PMN formation by fostering neutrophil N2 polarization. b The gut-liver axis consists of the intestinal epithelial barrier and the GVB, which collectively protect the liver from invasion by commensal or pathogenic microbes from the intestine. Alterations in the gut microbiota induce immune responses in the liver, mediated by the recruitment of MDSCs and Tregs, and reduction in Th17, NK cells, KCs, and NKT cell infiltration. This process is influenced by dysbiosis induced by various pathogenic bacteria and dietary factors. c Tumor-secreted factors directly or indirectly influence OC activation, leading to the formation of bone PMN. In BC bone metastasis, tumor-derived EVs containing miR-21 regulate the expression of programmed cell death 4, impacting PMN formation. Upregulation of circIKBKB significantly enhances IκBα phosphorylation, inducing the expression of GM-CSF and M-CSF, effectively promoting osteoclastogenesis. In HCC bone metastasis, large oncosomes (LOs) facilitate cellular cytoskeletal rearrangement and OC formation, while cytokines from HCC cells also contribute to bone metastasis. Additionally, EVs secreted by PCa induce premetastatic osteoblastic lesions, and cholesterol homeostasis in BM stromal cells plays a gatekeeping role in regulating PCa-promoting EV signal transduction. The immune TME plays a crucial role in bone PMN formation before bone colonization and is sensitive to various extracellular substances

Fig. 7

Chemical structures of small-molecule inhibitors discussed in the review. In this figure, we provide a comprehensive summary of the chemical structures of various small molecule inhibitors discussed in the review. These inhibitors are arranged in the figure according to the sequence in which they appear in the text

Fig. 8

Schematic illustration of the production process and influencing factors for TIL therapy. TIL therapy is a personalized treatment of cancer. After the patient’s tumor was removed, autologous T lymphocytes obtained directly from the surgically removed tumor were then cultured and expanded under IL-2 stimulation. The amplified TILs are then selected for recognition of autologous tumor cells and the resulting product is injected back into the patient, or directly by the Young TIL method, a method that does not require in vitro selection for tumor reactivity, and the TIL is rapidly amplified and injected back into the patient. Today, TIL therapy has shown remarkable clinical results in metastatic melanoma, advanced cervical cancer, and certain B-cell malignancies. Initial efficacy has also been achieved in NSCLC, CRC, and BC. However, patient and disease characteristics such as tumor metastasis, elevated serum lactate dehydrogenase levels, or unhealthy lifestyle habits, along with potential toxicities of IL-2 and physiological stress, limit its selective use, rendering patients with severe organ dysfunction, advanced age, or frailty ineligible for treatment. Moreover, despite achieving substantial objective responses, challenges include the polyclonal nature of TIL products with only a small subset being tumor-specific, as well as immune inhibitory mechanisms, which hinder effective tumor infiltration or full exploitation of TIL anti-tumor functions. Additionally, the association between tumor transcriptional characteristics, high tumor mutational load, neoantigen load and epigenetic modifications, and poor responsiveness to TIL treatment cannot be underestimated. Meanwhile, the experimental observation suggests the complexity of the underlying tumor-immune interactions and their importance in the TIL treatment process. Thus, the localization, aggregation, interaction with tumor cells and co-stimulation of all T lymphocyte subsets in the TME are necessary for a successful antitumor immune response