【Tomorrow at noon】

Cancer metastasis refers to the process in which cancer spreads from the origin of the primary tumor to farther parts of the body. The liver is a common metastasis site for many malignant tumors, and the incidence of liver metastasis diseases is 20 times that of primary liver cancer. The microenvironment, including vascular niches, is utilized by the tumor and is assimilated by the metastatic tumor microenvironment.

Angiogenesis is considered an important part of tumor growth and development. This key point is called the "angiogenesis switch". However, studies have shown that various tumors can grow and metastasize through non-angiogenic processes, suggesting that cancer cells can migrate and use the original blood vessels of the host organ to support tumor growth, survival and metastasis. We and others have discovered important prognostic and therapeutic significance for two different hepatic metastasis histopathology (HGP): connective tissue proliferation hgp (dhgp), which is a complete guide to the development of the tumor.

Vasi-vascularization is vascularized through angiogenesis and is responsive to anti-angiogenic drugs (bevacizumab (bev)) containing chemotherapy regimens; and alternative hgp (rhgp), which selectively obtains blood supply through vaso-selectively and is insensitive to bev. This suggests that vaso-co-selective may be a mechanism for initial and acquired resistance to anti-angiogenic treatment. A large number of patients need to be studied to further verify the role of hgp in drug resistance, as this will have a significant impact on the treatment of these patients.

Vascular selectivity is common in highly vascularized tissues such as the brain, lungs and liver. Importantly, the liver is the organ that is most likely to undergo selective metastasis, a phenomenon known as organ tropism. Therefore, we first discuss the unique characteristics of liver microstructure, which is a fertile soil for cancer cell growth, and then propose a theoretical model currently used to study vascular co-selecting, and then describe the different microenvironment components supporting co-selecting.

1. Unique features of liver microstructure

The liver is responsible for metabolism, clearance, biosynthesis and host defense. The liver's unique and diverse structures and cellular composition enable it to assume these special functions. The basic tissue unit of the liver is the lobular, which is a hexagonal prism tissue with a length of 2 mm and a diameter of 1 mm. Each corner of the prism has a portal vein triple consisting of bile ducts and two afferent blood vessels (hepatic artery and portal vein). In the center of the prism, an efferent drainage system, the "central" vein of the liver, collects the blood of the liver, through the lower

The vena cava reaches the body circulation. The blood flows from the afferent and efferent vessels from the afferent and efferent vessels to the hepatic vein in a centripetal direction. Unlike capillaries in other parts of the body, there are open fenestrated endothelial cells (ec) in the liver sinusoids, and the endothelial cells do not have a temporary basement membrane (bm), which separates from the underlying hepatic cells, allowing the plasma to have space to interact with the hepatic cells (disse gap) and hepatic stellate cells (pericellular cells of the liver). This unique function of bm allows molecules to inject into the disse space between the hepatic cells and the hepatic sinusoid.

2. Theoretical model of vascular co-selecting

The occurrence of liver metastasis is often described as a metastatic cascade, from the formation of pre-metastasis niches, microvascular invasion, extravascular formation, and then the vascularization and growth of tumor cells. The current theory suggests that successful metastasis proliferates, recruits endothelial cells, and activates the angiogenesis pathway to grow new blood vessels. We now know that this is not the only way to obtain the vasculature, and there are other options. A major pathway is vascular co-selecting, and by examining the histopathological phenotypes of these processes, we observe very different microenvironments (Figure 1). In this review, we will focus on two major vascularization types already documented in CRCCLM: angiogenesis and co-selecting.

2. Theoretical model of vascular co-selecting

Our theory is that blood vessels are selected in 4 different stages, which are flowing and can be performed simultaneously.

—The first stage, motility, requires that cancer cells be able to move so as to get close to the hepatocytes, thus occupying the space occupied by the hepatocytes and selecting sinusoids for the blood. Together with our colleagues, we have shown that if we reduce the motility of cancer cells, the proportion of lesions using angiogenesis will increase, and the exact mechanism that causes this phenomenon has not been elucidated, but there is new evidence that we review this in Section 1.

—Stage 2, shifting, i.e., cancer cells transfer hepatocytes and occupy their space for entry into sinusoidal blood vessels. How this shift occurs is unclear and may include physical shifting, creating an environment that promotes apoptosis, necrosis, or autophagy of hepatocytes. In addition, liver cells may be transformed by factors secreted by cancer cells (i.e. extracellular vesicles) that may change their molecular behavior. We have evidence that at least a portion of hepatocytes are shifted and bound to the front of the tumor by one or several of the methods mentioned above. Our current work shows that co-selected cancer cells not only come into close contact with the liver cells, but also entangled with each other. The micrograph in Figure 3 shows a hepatocyte-specific antigen (HSA) stained vascular proliferative CR

Clum lesions, showing that hepatocytes at the edge of the lesion are mixed with cancer cells, actually found some in the center of the lesion. Apart from the fact that they are replaced, it is interesting that despite the fact that these cells have decisively stained HSA, they do not necessarily possess all the characteristics of normal hepatocytes, and of course their polarity has been disrupted, suggesting that they may be being altered/transformed into other types of cells. There is no hSA staining in the angiogenesis (dhgp) lesions on the side of the proliferative ring of connective tissue, which we have only seen in proliferative (rhgp) lesions. As for the possibility that apoptosis and/or autophagy play a role in this displacement, there is no evidence of support or objection so far.

——Stage 3, interaction and regulation of cancer cells and vascular system. Once cancer cells occupy the position of hepatocytes, they can enter the hepatic vascular system, mostly hepatic sinusoids, but may also include triplet arteries and portal veins. In the third stage, cancer cells interact and regulate this vascular system, one situation is to promote angiogenesis, and the other situation is to inhibit angiogenesis and promote vascular selection. The vascular components that need to be manipulated include extracellular matrix, adhesion pathways, endothelial angiogenesis and angiopoietin-bonding pathways, and stellate cells. This is further described in Section 3.

——Stage 4, recruitment of immune components of vascular co-selected microenvironment, we observed recruitment of immune components unique to vascular co-selected microenvironment described in Section 4. It should be emphasized that the vascular properties of liver metastases are a "fluid" process, and these tumors can transition from angiogenesis to co-selecting, and vice versa. We believe that the interaction of microenvironment components (described here) is a key driver and promoter of this process in different stages.

? Establish a proliferative Crclm lesion model. Schematic diagrams showing the development of angiogenesis and colorectal cancer liver metastatic lesions. In selective lesions, due to overexpression of arp2/3, cancer cells show higher motility than angiogenic cells. Therefore, cancer cells invade the liver plate and replace liver cells without destroying the liver vascular structure at the tumor-liver junction, which is conducive to the selectivity of preexisting sinusoidal blood vessels. Schematic diagram of microenvironment changes accompanying the development of vascular enlargement.

3. The role of cancer cells in vascular selection

1) Tumor cell motility

Since vascular co-selecting requires cancer cells to penetrate into the normal liver parenchyma, inducing cancer cell movement is the first stage of establishing vascular co-selecting, which allows cancer cells to invade their adjacent tissues and co-select existing blood vessels.

Different molecules such as arp2/3, cxcr4, cxcl12, oleg2-wnt7a and thbs1 are involved in supporting the movement and infiltration of cancer cells to support vascular selectivity.

2) Tumor cell adhesion

In vascular selection, tumor spread to vascular accompanies the adhesion of cancer cells to the basement membrane of the vessel or pre-existing vascular endothelial cells. Integrin molecules include α3 integrin, α6 integrin and β1 integrin, which promotes adhesion of cancer cells to blood vessels. Inhibition of the above integrin may be an easy-to-operate strategy to destroy resistance to anti-angiogenesis treatment. l1cam is a cell surface glycoprotein that is highly expressed in metastatic tumors. Other studies have also confirmed that l1cam is involved in the adhesion of cancer cells to the original blood vessels of metastatic tumors. It has also been reported that l1cam prevents the maturation of tumor blood vessels, and the inhibition result is to induce vascular normalization, which is believed to improve tumor response to cytotoxic drugs. Therefore, pharmacological strategies for l1cam signaling may potentially eliminate vascular selectivity.

3) Epithelial interstitial transformation of proliferative cancer cells (emt)

Epithelial-to-mesenchymal transformation (EMT) is the process by which epithelial cells obtain a mesenchymal phenotype, and plays a key role in embryogenesis, wound healing, fibrosis and cancer. Previous studies have strongly demonstrated the correlation between the emt process and tumor vascular selection. Therefore, further studies are needed to reveal the underlying molecular mechanism of the role of emt in tumor vascular selection.

4) Metabolic reprogramming in cancer cells

Cancer cells have been well proven to be metabolicly active, and they rely on angiogenesis to ensure a sustained supply of oxygen and nutrients. Metabolic signaling pathways are closely related to angiogenesis signaling events. It is observed that cancer cells have higher glycolysis than normal cells, which leads to the hypothesis that oxophosphos is generally downregulated in cancer. Therefore, it is conceivable that angiogenesis switches and metabolic switches go hand in hand.

4. Extracellular vesicles produced by cancer cells (evs)

Since vascular co-selecting requires interactions between cancer cells and the extracellular environment, cancer-derived extracellular vesicles (evs) can promote tumor co-selecting microenvironments in specific niche environments. evs from cancer cell-derived evs play the role of the main coordinator in angiogenesis and vascular entropy, and they play a crucial role in the third and fourth stages of vascular co-selecting development. Therefore, taking into account all observations, cancer cells utilize various mechanisms to develop vascular selectivity (Figure 4).

5. Extracellular matrix (ecm) components

The tumor microenvironment includes many different types of cells except cancer cells. The extracellular matrix (ECM) is an important part of the tumor microenvironment. ECM forms a complex network that mainly contains water, proteins and proteoglycans. Its role is to provide physical scaffolds and structural support for cells. By interacting with cell surface receptors and matrix components, the extracellular matrix provides surfaces for cell adhesion, proliferation, growth, and serves as a reservoir of signaling molecules. In fact, it is able to bind and secrete growth factors and cytokines to promote morphogenesis, metabolism and cellular functions involved in angiogenesis. Therefore, the ECM network includes a complex microenvironment that is highly dynamic in nature and can undergo continuous remodeling.

6. Vascularization of liver metastases

"How do we identify tumor angioselective and angiogenesis-driven tumors?" Some studies have shown that non-angiogenetic tumors have a unique morphology.

1) Hepatic Vascular System and Molecular Biology The liver is a highly vascularized organ rich in oxygen and nutrients. Metastatic cancer cells reaching the liver can initially survive the spread of oxygen and nutrients, and eventually they must be vascularized to increase volume. Obtaining a blood supply that can provide oxygen and nutrients is crucial for the progress of all types of cancer. The construction of liver blood vessels is very unique, characterized by a sinusoidal network of blood vessels. Cancer cells must occupy the space occupied by the liver cells in order to use the liver sinusoidal supply blood. An interesting feature of all proliferative tumors is that they retain the original tissue structure, and angiogenic tumors usually destroy the original normal components of the infiltrating organs. Therefore, a "partnership" is formed between cancer cells and liver sinusoidal endothelial cells (lsec).

2) Blood circulation is a dynamic system. A lot of work has been done on the mechanism of angiogenesis, including the growth of endothelial cells, stimulating the degradation of the basal membrane of the blood vessels, and the loss of pericytes associated with blood vessels. Recently, a large number of papers have begun to study the mechanism of co-selectivity. Joint selection has been studied and is associated with worse tumor progression. In clinical models of lung, glioblastoma, hepatocellular carcinoma, and melanoma brain metastasis, as the mechanism of initial and acquired resistance to anti-angiogenesis treatment (anti-vascular endothelial growth factor (vegf)), clinically, in glioblastoma patients, we have also received support from our Crclm patients.

7. The role of immune cells in vascular co-selecting

It is worth noting that the il-8/cxcr1 pathway has been shown to be the promoter of non-angiogenic vascular mimics in glioblastoma. In Crclm lesions, Hand et al. observed higher levels of il-10, il-6 and VEGF, while the levels of il-10 in adjacent livers are also elevated compared to healthy livers. Whether these cytokines are differentially expressed in angiogenic tumors and proliferative tumors has not been studied. There is a growing body of evidence that tumor-invasive immune cells affect angiogenic. Some studies have shown that angiogenic switches are immune switches caused by pro-angiogenic polarization of immune cells. Therefore, understanding the "immune environment" and focusing on defining adaptive or innate immune responses may help define angiogenic tumors and tumor selection, which plays a key role in the fourth stage of vascular entropy, as described by our theoretical model.
Chapter completed!