Colorectal cancer (CRC) is one of the most common malignancies for both males and females, ranking in the top three for both estimated incident cases and deaths (1). WNT signaling hyperactivation induced by mutational inactivation of the Adenomatous Polyposis Coli (APC) accounts for most CRC cases in patients (2). In addition, other somatic mutations, such as the tumor suppressor P53 and oncogenic pathway KRAS are commonly observed in CRC (3, 4). Although these tumor cell-autonomous pathways significantly contribute to CRC progression and metastasis, therapeutic interventions that target these pathways have achieved limited success in patients. Chemo- and radio-therapies, which have substantial adverse effects, are still commonly used for CRC patients, especially for those at advanced stages. Therefore, therapeutic strategies with better efficacy and less toxicity are urgently needed.

Accumulating evidence indicates that non-cell-autonomous pathways, especially signaling pathways of the tumor microenvironment (TME), are significantly involved in CRC progression and metastasis, either by promoting or inhibiting the process. TME refers to a special biological environment formed by malignant cells, non-malignant cells and their secreted components as summarized in a previous review (5). Given the pivotal roles of the TME in CRC progression and metastasis, exploration of the mechanisms underlying the interplay between TME remodeling and CRC development have attracted substantial attention over the last decade. Great progress has been made in this field, which has dramatically advanced our knowledge about the TME and CRC. New insights about CRC treatment have also been elucidated. However, a complete understanding of the TME in CRC progression and metastasis has yet to be unfolded.

The TME dynamically changes with enormous complexity throughout cancer progression. In general, it is composed of cellular and non-cellular components, which have distinct functions but also collaborate with each other during cancer progression and metastasis (6). In this review, we summarize our current understanding of the roles of the TME in CRC progression and metastasis by focusing on the components in both primary sites and distant metastatic organs. We also briefly discuss the therapeutic insights on remodeling the TME in CRC.

The main components of the TME are the extracellular matrix (ECM) and its cellular partners, including immune cells, endothelial cells and fibroblasts. In addition to these common components, intestinal cells are also in close contact with a large population of microorganisms referred to as the gut microbiota (7). In this section, we summarize the roles of these common components as well as a few key microorganisms in CRC progression and metastasis.

The critical roles of microorganisms in our gastrointestinal tract have long been known. On one hand, the microorganisms help with digestion and maintain homeostasis. On the other hand, they establish and promote disease progression (Figure 1). One of the most famous examples is Helicobacter pylori (H. pylori). The positive correlation between H. pylori infection and gastric cancer has been well recognized (8). Compared to the stomach, the colon has an even more diverse microbiota, which is also significantly involved in maintenance of homeostasis and progression of diseases.

In addition to microorganisms, the primary CRC tumors have cellular components within its microenvironment. In this section, we discuss the tumor promoting or suppressing roles of immune cells, cancer-associated fibroblasts, and endothelial cells (Figure 2).

In addition to the cellular components discussed above, the ECM built by the cellular components serve as a scaffold and is an essential component of the TME. Small molecules and vesiculas secreted by these cells facilitate signaling transduction and play essential roles in CRC progression and metastasis.

During late-stage development of CRC, tumor cells metastasize to distant organs. The tumor cells together with stromal cells create a microenvironment to either foster or restrain the outgrowth of the metastatic tumors. As such, the TME in metastatic organs harbors common cellular and non-cellular components as discussed above. However, the metastatic organs could also have their own unique stromal components that are involved in CRC colonization and outgrowth (Figure 3).

Around 20-25% of patients initially diagnosed with colorectal carcinoma present with liver metastases (100). Of these, 80–90% have multiple organ metastasis, including liver, and 50% have exclusively liver metastasis (101). This is partially due to the liver being proximal to the colon and their intimate connection by portal circulation. A large number of tumor cells upon extravasation from primary sites could directly disseminate to liver via blood circulation. Meanwhile, the liver may create a unique microenvironment to attract and foster tumor colonization and outgrowth. To illustrate, Kupffer cells, which are a set of specialized cells localized in the liver, highly express CXCL12 to attract CXCR4-expressed CRC cells. This facilitates tumor cell colonization and outgrowth (102). Likewise, liver parenchymal endothelial cells (LPECs) activate the HER3-AKT pathway in tumor cells via paracrine signaling to promote CRC liver metastasis (103). Conversely, liver cells could also restrain metastatic colonization. For example, liver resident NK cells provide immune surveillance and eliminate tumor cells that disseminate to the organ. Harmon et al. found that reduction of these liver resident NK cells significantly promotes CRC liver metastasis (104).

Pro- or anti-metastasis communication between CRC cells and liver stromal cells could occur even before the residence of tumor cells in the liver. Tumor cells reshape the liver microenvironment through secretomes before their arrival. Shao et al. found that CRC secreted exosomes are enriched with microRNA-21-5p (miR-21). These induce liver macrophage polarization toward an interleukin-6 (IL-6)-secreting proinflammatory phenotype. As a result, the proinflammatory niche promotes liver metastasis (105). However, exosomal Angiopoietin-like protein 1 (ANGPTL1) attenuates CRC liver metastasis. Mechanistically, exosomal ANGPTL1 is taken up by Kupffer cells in the liver, which alters their secretion patten by targeting the JAK2-STAT3 pathway. The altered Kupffer cells decrease MMP9 expression and prevent liver vascular leakiness, and consequently, suppress liver metastasis (106).

The lungs are the second most common metastatic site of CRC after the liver. Commonly, lung metastasis occurs with an incidence of 10–15% in patients after radical resection of CRC (107). Previous study found that the activation of NF-κB signaling in CRC cells enhances TNF-α production of host hematopoietic cells. This results in a pro-inflammatory microenvironment within the lungs that promotes CRC lung metastasis (108). Similar to the liver, the secretomes of tumor cells could remodel the lung microenvironment to favor CRC lung metastasis. Exosomal miR-25-3p secreted by CRC cells facilitates metastasis by promoting vascular permeability and angiogenesis by targeting endothelial cells in the lung. Mechanistically, it targets KLF2 and KLF4 to regulate the expression of VEGFR2, ZO-1, occludin and Claudin5 (109). In addition, CRC primary tumors release integrin beta-like 1 (ITGBL1)-rich extracellular vesicles (EVs) to stimulate the TNFAIP3-mediated NF-κB signaling pathway and activate resident fibroblasts in distant organs (110). These activated fibroblasts produce pro-inflammatory cytokines such as IL-6 and IL-8 to create a pre-metastatic niche in the lung (110).

In addition to the liver and lungs, CRC cells could metastasize to the peritoneum, brain, and ovary. Five to ten percent of CRC patients are diagnosed with peritoneal metastases. This increases to 20-50% in recurrent CRC patients (111). Peritoneal metastasis can originate from preoperative tumor cell dissemination, intraoperative trauma-induced tumor cell shedding, metastatic lymph node rupture, lymphatic tumor thrombus rupture, or surgical field hemorrhage. All of which induces the implantation of free cancer cells in the peritoneum (112, 113). The peritoneal microenvironment is composed of a variety of cellular components, such as human peritoneal mesothelial cells (PMCs), peritoneal fibroblasts (PFBs), peritoneal macrophages (PMs), and adipocytes, which create a suitable soil for peritoneal metastasis. For example, the expression of intercellular adhesion molecule-1 (ICAM-1) and vascular adhesion molecule-1 (VCAM-1) in CRC cells can increase their adhesion to PMCs (114).

Two to twelve percent of CRC patients develop brain metastasis (115). Significantly, the blood-brain barrier is a natural fence that separate the brain with the rest of the body. It creates a unique brain microenvironment. It also serves as the first line of defense toward tumor cells dissemination to the brain. Nevertheless, CRC tumor cells can still metastasize to the brain. Previous studies revealed that nitric oxide (NO) is a crucial mediator of anti-tumor properties of microglia in the TME of the brain (116). However, metastatic CRC cells may suppress the cytokine-induced NO production in cerebral endothelial cells to inhibit the activation of microglia (117), and consequently, promote the outgrowth of CRC brain metastases.

Ovarian metastasis of colorectal cancer is relatively rare and usually occur in younger patients (118). Notably, the ovaries are rich in lymphatic vessels and there are lymphatic junctions between the colorectum and ovaries bilaterally. Studies suggested that CRC cells take advantage of the lymphatic vessel rich microenvironment to invade the lymphatic system and subsequently metastasize into the ovaries (119, 120).

Thanks to the development of cancer screening and therapeutic strategies, the number of CRC deaths has been declining over the last few decades. However, at least 50,000 patients may die from CRC in 2022 within the United States. It is still a devastating disease, and curative therapeutic interventions are urgently needed. Given the critical functions of the TME in CRC progression and metastasis and our accumulating knowledge on this subject, new insights about CRC therapy by targeting TME are emerging (Figures 1, 2).

As discussed above, the gut microbiota plays essential roles in CRC progression and metastasis. Manipulation of the gut microbiota could hold clinical implications for CRC patients. Intratumor Gammaproteobacteria residing in the TME metabolize the chemotherapeutic drug gemcitabine (2′,2′-difluorodeoxycytidine) into its inactive form, 2′,2′-difluorodeoxyuridine. Antibiotic-induced depletion of this bacteria enhanced the gemcitabine-mediated chemotherapy response in a CRC mouse model (121). Alternatively, fecal microbiota transplantation (FMT) may also be employed to prevent CRC in patients with dysbiosis (122). A FMT procedure could help eradicate procarcinogenic E. coli (114). Encouraged by these pre-clinical findings, a few clinical trials on manipulating gut microbiota in CRC patients are ongoing (123).

Immunotherapies such as immune checkpoint blockade therapy, Chimeric antigen receptor (CAR) T-cell therapy, and T-Cell Receptor (TCR) Therapy have achieved considerable success in several cancer types (124). Immune checkpoint blockade therapy such as PD-1 and CTLA-4 inhibitors have been approved as first or second line of treatment for metastatic CRC patients with microsatellite instability-high (MSI-H). Unfortunately, treatments in MSI-H CRC patients usually only extend survival rate rather than induce complete regression of metastatic lesions. Even worse, the majority (>85%) of CRC patients with microsatellite instability-low (MSI-L) are not eligible for this therapy. Immunotherapy in combination with other conventional treatments, such as chemotherapy, radiotherapy, and targeted therapy are currently being tested in order to overcome these obstacles and to encompass a broader CRC population with better efficacy (125).

Given the central prooncogenic roles of CAFs, therapies that target CAFs have been emerging. Fibroblast activation protein (FAP) is a commonly used marker for CAFs and serves as an ideal target for CAFs eradication. Pre-clinical studies indicate that administration of DNA vaccine that targets FAP effectively depletes CAFs in the TME, and consequently suppresses CRC progression and metastasis (126). Furthermore, CAF depletion by small-molecule dipeptidyl peptidase inhibitor PT-100 significantly enhanced the efficacy of chemotherapy in CRC mouse models (127). In addition, a small chemical compound, Atractyloside, was also found to inhibit CRC metastasis by targeting CAFs (128). Collectively, all these preclinical studies suggest that CAFs in the TME could be a potential target for CRC.

One of the most common therapies for CRC is to target VEGF and prevent endothelial cell-mediated angiogenesis. However, this method has considerable adverse side effects and limited benefit to patients because it targets both TECs and normal endothelial cells. Nanoparticles that could selectively recognize and deplete TECs have been developed and tested in pre-clinical models (129).

Aside from targeting or remodeling cellular components in the TME, manipulating non-cellular components could have therapeutic potential as well. As mentioned above, fibronectin is one of the major components of the ECM that promotes CRC progression and metastasis. ATN-161, which is a non-RGD-based pentapeptide (PHSRN) derived from the fibronectin synergy region, inhibits breast cancer progression and metastasis by antagonizing fibronectin (130). It was shown that ATN-161 suppresses metastatic CRC progression and sensitize it to chemotherapy in mouse model (131). Similarly, HM-3, which is another integrin RGD domain containing peptide, remodels the ECM and achieves impressive anti-tumor efficacy in CRC xenograft models (132). In addition to remodeling the ECM, manipulating other cell-secreted TME components have attracted attention. Ascites-derived exosomes (Aexs) together with the granulocyte-macrophage colony-stimulating factor (GM-CSF) achieved encouraging outcomes in a phase I clinical trial with advanced CRC patients (133).

The TME has been recognized as a key player in tumor progression and metastasis in CRC. All the components from both microorganisms and the host are significantly involved in this process. Although each component has distinct roles during CRC progression and metastasis, most of them behave like a double-edged sword, which could both promote or inhibit tumor progression depending on the specific context. Moreover, the TME is highly dynamic during CRC progression and metastasis. The properties and abundance of each TME component could be significantly altered during tumor progression and treatments. In this review, we only covered a few common components and discussed their functions mainly from a tumor-promoting point of view. Of note, polymorphic microbiomes are one of the hallmarks of cancer. Many organs other than the colon, including the liver and lungs may have distinct microbiomes (134). Here, we only discussed the gut microbiota and their potential function in primary tumor sites. However, as mentioned above, both microorganisms and host non-malignant components dramatically contribute to the initiation and progression of CRC.

Cancer metastasis is composed of a serial of complex process, from extravasation from primary sites to the eventual colonization at distant organs (135). Although a considerable number of CRC patients have metastatic disease at the time of diagnosis or recurrency, only a small fraction of the disseminated cancer cells are capable of successful metastases (136). Disseminated tumor cells encounter tremendous stresses from the TME during metastasis (137). Each component of TME contributes to this metastatic cascade. The specific components of the TME and their roles during the metastatic cascade have not been covered here as it has been comprehensively discussed in our previous review (137). In this review, we mainly focused on established metastases. Moreover, we emphasized unique metastatic organ components rather than restated the common TME components (such as CAFs and ECM) of primary tumors. Collectively, unique stromal cells of various tissues promote or restrain metastatic outgrowth with distinct pathways as discussed above.

With the accumulating knowledge about TME in CRC progression and metastasis, promising therapeutic strategies that modulate TME are emerging. As mentioned earlier, many researchers have confirmed that better anti-tumor effects can be achieved through remodeling of the TME. In addition to the beneficial effects in pre-clinical studies, observations in clinical trials have shown the great potential of TME remodeling to improve the therapeutic effect of drugs.

However, due to the dynamic changes of the TME during cancer progression and metastasis, a detailed, functional dissemination of every TME component at each given tumor stage is essential. This knowledge would help us stratify CRC patients that could benefit from a specific TME remodeling therapy, and more importantly, suggest the best treatment window. Furthermore, monotherapy is usually insufficient because tumors, especially metastatic tumors, have the tendency to develop resistance. In order to minimize treatment resistance, combining TME remodeling strategies with other promising therapeutics, such as immunotherapy, is another aspect we need to explore in future studies.

JL and MS designed and drafted the manuscript. DC participated in revising the manuscript critically. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.