Interplay of genetic and clinical factors in cancer-associated thrombosis: Deciphering the prothrombotic landscape of colorectal cancer

Abstract

Colorectal cancer (CRC), the third most prevalent cancer globally, exhibits a notable association with venous thromboembolism (VTE), significantly impacting patient morbidity and mortality. We delve into the complex pathogenesis of cancer-associated thrombosis (CAT) in CRC, highlighting the interplay of clinical risk factors and tumor-specific mechanisms. Our comprehensive review synthesizes the current understanding of CRC’s pro-thrombotic tendencies, examining both general clinical factors (e.g., age, gender, obesity, prior VTE history) and tumor-specific aspects (e.g., tumor location, stage, targeted therapies). Key findings illustrate how CRC cells themselves actively contribute to coagulation cascade activation through various procoagulant elements such as tissue factor, cancer procoagulant, and extracellular vesicles. We also explore how CRC influences host cells to adopt a procoagulant phenotype, thereby exacerbating thrombotic risks. This review underscores the role of genetic mutations in CRC (e.g.,KRAS, p53) in modulating coagulation-related protein expression and thrombosis risks. An in-depth understanding of the genetic landscape specific to CRC subtypes is essential for developing targeted anticoagulation strategies and could significantly advance thrombosis prevention while improving the overall management of patients with CRC. This highlights the urgent need for precision in addressing CAT within clinical settings.

Key Words: Colorectal cancer cell; Cancer-associated thrombosis; Venous thromboembolism; Tissue factor; Platelet

Core Tip: This comprehensive review examines the complex interplay between genetic and clinical factors in colorectal cancer (CRC)-associated thrombosis. It highlights how tumor-specific mechanisms, such as tissue factor and cancer procoagulant, actively contribute to the coagulation cascade, alongside clinical risk factors like age, gender, and obesity. By integrating both tumor biology and patient-specific characteristics, this study provides crucial insights into the prothrombotic tendencies of CRC, offering valuable perspectives for personalized therapies and improving patient management in CRC-associated venous thromboembolism.

INTRODUCTION

As of 2020, colorectal cancer (CRC) ranks as the third most common cancer globally and produces the second-highest mortality rate among cancers[1]. The intriguing association between idiopathic thromboses and concealed malignancies was first proposed in 1865 by Armand Trousseau, who identified superficial thrombophlebitis as a potential early marker of hidden visceral cancers[2]. Notably, CRC exemplifies this connection, with reports citing CRC as presenting one of the highest incidences of thrombotic events among various prevalent cancers[3]. Moreover, hypercoagulable states in patients with CRC have been associated with both clinical progression and prognosis.

The intricate association between cancer and venous thromboembolism (VTE) is widely acknowledged, although its etiology remains a subject of ongoing research. Notably, about 20% of patients with VTE are concurrently diagnosed with cancer, and patients with cancer exhibit a substantially heightened risk of VTE, ranging from 14% to 25%[4-6]. This risk is most pronounced during the initial three months following a cancer diagnosis[7,8]. Moreover, cancer-associated thrombosis (CAT) significantly contributes to both morbidity and mortality, with VTE ranking as the second leading cause of death in patients with cancer, following the cancer itself[9,10]. Patients with metastatic cancers, in particular, have a higher thrombosis risk than those with non-metastatic forms. Even in the absence of overt thrombosis, most patients with cancer exhibit hypercoagulability, detectable through laboratory testing[4]. Moreover, cancer treatments, including surgery and chemotherapy, are known to elevate the VTE risk[11,12].

The pathogenesis of CAT in CRC is complex and multifaceted. Clinical factors encompassing general risks such as advanced age, being a woman, obesity, prior VTE, infection, and surgery; along with tumor-specific factors like tumor location, advanced stage, and metastatic disease, contribute to the thrombotic risk in patients with CRC. Moreover, targeted vascular therapy is a notable risk factor. CAT pathogenesis involves diverse mechanisms. For instance, tumor cells can activate the host’s hemostatic system in multiple ways, often driven by oncogenes implicated in neoplastic transformation[13]. These tumor tissues express an array of procoagulant proteins, including tissue factor (TF), cancer procoagulant (CP), and various coagulation factors. They also shed procoagulant-associated extracellular vesicles (EVs) and can induce procoagulant properties in host cells. This induction occurs either through direct cell-cell adhesions or by releasing inflammatory cytokines and proangiogenic factors.

In this article, we examine the progress in understanding the thrombogenicity of CRC, with a special emphasis on the mechanisms underlying tumor-specific activation of procoagulant properties.

CLINICAL RISK FACTORS FOR CANCER-RELATED THROMBOSIS IN CRC

Clinical risk factors for thrombosis in CRC encompass both general clinical and tumor-specific factors. These factors are prevalent in all sorts of patients with and without cancer. However, the presence of tumor-specific clinical risk indicators and biological markers in patients with malignant tumors lends a unique aspect to the pathogenesis of thromboses associated with CRC. The interplay of these factors results in a shift of the hemostatic balance towards a hypercoagulable state that differentiates the thrombotic process in CRC from other conditions (Figure 1).

Figure 1 Clinical risk factors influencing thrombosis in colorectal cancer. This figure delineates two primary categories of clinical risk factors: General clinical factors and tumor-specific factors. General clinical factors, prevalent in both tumor and non-tumor patients, contribute to the overall risk landscape. However, it is the presence of tumor-specific clinical risk indicators and biological markers unique to malignant tumors that particularly delineates the pathogenesis of thrombosis associated with colorectal cancer. Collectively, these factors precipitate a shift in hemostatic equilibrium, favoring a hypercoagulable state. Created in BioRender (Supplementary material).

General clinical factors

General clinical factors are pertinent to both oncological and non-oncological patients. These factors encompass a range of conditions, including age-related risks (particularly in elderly patients), a history of thrombosis and embolism, obesity, infection, surgical history, anemia, and prolonged immobilization post-surgery. Notably, sex is an independent risk factor for CAT in patients with CRC. A retrospective study by a Japanese scholar suggested that being women is an independent risk factor for preoperative VTE in CRC[14]. In addition, a large-scale, multicenter, prospective cohort study in China showed that being a woman was an independent risk factor for postoperative VTE in CRC[15]. Possible reasons for this include the fact that estrogen increases levels of coagulation factors and decreases fibrinolytic activity. Moreover, higher levels of inflammatory cytokines such as interleukin (IL)-6 and tumor necrosis factor (TNF-α) in women may activate the coagulation system, leading to an increased risk of thrombosis[16]. Laboratory indicators of thrombotic risk, such as elevated white blood cell and platelet counts, specific ABO blood group types, increased D-dimer levels, and a heightened maximum platelet aggregation rate, also fall under this category. Patients with CRC often exhibit raised platelet and neutrophil counts, a trend associated with an elevated thrombosis risk and a poor cancer prognosis[17-19]. The Khorana risk model, widely used in oncology, integrates these factors-platelets, leukocytes, tumor site, hemoglobin, and body mass index-to predict the CAT risk[20]. The model’s efficacy has been corroborated by multiple extensive studies[21-23]. Moreover, non-O blood groups have been associated with an increased VTE risk in patients with CRC[24]. Patients with a genetic predisposition for thromboses also present an increased risk of VTE. Major genetic defects have been identified, including loss-of-function mutations in natural antithrombin, protein C, and protein S, as well as gain-of-function mutations in procoagulant factors V (FV Leiden) and II (prothrombin G20210A)[25]. The risk of each genetic defect by itself may be relatively low, but the coexistence of several defects may significantly increase susceptibility to thrombosis. In addition, environmental factors may interact with one or more of the genetic variants to further increase the risk.

Tumor-specific factors

Tumor location is a predominant tumor-specific factor. CRCs on the left side (encompassing rectal, sigmoid, and descending colon cancers) demonstrate a higher propensity for thrombosis than right-sided cancers (including ascending and transverse colon cancers). This tendency is attributed to higher blood viscosity, slower blood flow, and increased intestinal inflammation associated with left-sided CRC tumors, which may promote thrombosis[26,27]. Epidemiological data indicate that certain malignancies, like brain, hematological, pancreatic, gastric, and ovarian cancers, exhibit higher VTE risks than others[28,29]. Cancer stage is another crucial risk determinant, with advanced metastatic diseases posing a greater VTE risk than localized tumors[29]. In patients with advanced CRC, increased tumor burden and systemic inflammatory response promote the expression of procoagulant molecules[30]. Advanced tumors can obstruct or compress blood vessels, leading to stagnant blood flow, an important component of the Virchow’s triad that leads to thrombosis[31]. In addition, the presence of central venous catheters for long-term chemotherapy or parenteral nutrition in patients with advanced cancer can disrupt normal blood flow and lead to thrombosis[32]. The period immediately following a cancer diagnosis is characterized by a high VTE risk[7] probably due to the procedures and treatments common during this time: Diagnostic procedures, including colonoscopic biopsies, can lead to endothelial damage, generating a prothrombotic state and increasing the VTE risk. Many patients undergo surgical procedures such as colonoscopy or tumor resection after being diagnosed with CRC. In addition, treatments such as preoperative neoadjuvant chemotherapy may be initiated soon after diagnosis, further increasing the VTE risk.

The pathological tumor type may also influence the CAT risks. Mucinous CRC secretes excessive mucin as one of its distinguishing features. Studies have shown that TF-free oncolytic mucin preparations induce the formation of platelet-rich microthrombi in mice in a P-selectin-dependent and thrombin-independent manner[33,34]. Two other studies have also suggested that circulating Mucin1 (MUC1) attachment to EVs and other soluble mucins may also have a key role in thrombus formation[35,36]. However, the most recent study by Kawano et al[37] showed similar MUC1 levels in patients with cancer with or without VTE, and the data did not reveal any evidence of MUC1 being associated with VTE in patients with cancer. The conflicting evidence may be due to differences in mouse thrombosis models and the fact that Kawano et al[37] used a nested case-control study with a small number of patients, which led to results that are not generalizable. In the future, larger prospective cohort studies are needed to compare the role of mucins in thrombosis in tumors such as CRC. Therefore, the procoagulant mechanism of mucins in CRC needs further investigation.

Antitumor interventions, including chemotherapy, radiotherapy, antiangiogenic medication, and cancer surgery, can exert prothrombotic effects[38]. Chemotherapeutic agents and tumor derivatives compromise the vascular endothelial integrity, diminishing its antithrombotic properties. They also induce overexpression of TF and amplify cell membrane phosphatidylserine (PS) exposure[39]. Specifically, for chemotherapeutic agents, the oxidative stress caused by 5-fluorouracil can directly activate platelets, promoting thromboses[40]. Oxaliplatin-induced peripheral neuropathy can cause mobility problems, which are an important risk factor for VTE. In addition, oxaliplatin induces the release of pro-inflammatory cytokines and chemokines (e.g., IL-8, MCP-1), which attract leukocytes and promote a pro-thrombotic environment[41]. These chemotherapeutic agents also stimulate the release of TF-positive EVs (TF+ EVs) from tumor cells and endothelial cells (ECs), promoting coagulation[42]. Tamoxifen therapy against breast cancer can also cause VTE, likely because it decreases the levels of anticoagulant proteins, enhances platelet aggregation, and affects lipid metabolism[43]. Bevacizumab, a targeted vascular therapeutic agent for CRC, has also been associated with an increased thrombosis risk in oncology patients[44-46].

MECHANISMS UNDERLYING PRO-THROMBOTIC FORMATIONS IN CRC CELLS

Inherent procoagulant properties of tumor cells

CRC cells actively contribute to the activation of the coagulation cascade. This activation is primarily driven by the expression of various procoagulant elements. Key among these are TF, CP, and a range of tumor-derived EVs. Additionally, CRC cells exhibit elevated levels of various coagulation factors and fibrinolytic proteins, further exacerbating the pro-thrombotic environment. Figure 2 shows a depiction of this complex interplay.

Figure 2 Self-promoting procoagulant properties of tumor cells: An insight into tumor-induced coagulation mechanisms. This figure illustrates the multifaceted role of colorectal cancer cells in activating the coagulation system. Central to this process is the self-expression of tissue factor (TF), a pivotal protein in cancer-associated thrombosis. The regulation of TF is influenced by key genetic alterations, including activation of the KRAS oncogene and loss of the p53 tumor suppressor, as well as by the tumor’s inflammatory milieu. TF not only initiates coagulation via its complex with activated factor VII, leading to exogenous pathway activation, but also enhances tumor progression by upregulating vascular endothelial growth factor (VEGF) in both malignant and host vascular cells. Concurrently, cancer procoagulant activates factor X, bypassing the need for factor VII. The role of acetylheparinase in increasing TF activity through the dissociation of tissue factor pathway inhibitor is also noted. Moreover, we observe an upsurge in TF-positive extracellular vesicles (TF+ EVs) in colorectal cancer, reinforcing the coagulation cascade. The association of plasma coagulation factors VII (FVII) and VIII (FVIII) with hypercoagulability is highlighted, along with the thrombotic implications of Bevacizumab treatment, a VEGF-targeting drug that elevates plasminogen activator inhibitor-1 levels. PL: Phospholipid; VTE: Venous thromboembolism. Created in BioRender (Supplementary material).

TF is widely regarded as a central player in CAT due to its significant roles in both tumor progression and VTE. As a primary initiator of blood coagulation under both normal and pathological conditions, TF, in conjunction with activated factor VII (FVII), initiates blood coagulation via the extrinsic pathway. Its expression is a common feature of numerous solid tumors and hematological malignancies. The TF expression level is often directly proportional to a tumor’s aggressiveness. Empirical evidence, particularly in colorectal and pancreatic cancers, suggests a correlation between plasma TF levels and tumor size[30,47]. Further, studies focusing on human CRCs indicate that TF positivity is associated with clinical stage, histological grade, a poor prognosis, and angiogenesis[48-50]. These findings underscore TF’s role not only in cancer-associated coagulopathies but also as a potential indicator, and possibly a determinant, of malignant tumor cell behavior. In lung cancers, particularly in non-small cell lung cancers, mutations in the epidermal growth factor receptor (EGFR) pathway upregulate TF expression and promote coagulation[51]. In addition, hypoxic lung tumors express hypoxia-inducible factors, and these in turn upregulate procoagulant factors, which may differ from those in CRC[52]. Additionally, the high inflammatory status of patients with CRC further boosts TF production[53]. The levels of inflammatory cytokines such as IL-6, TNF-α, and IL-1β are relatively lower in CRC than in pancreatic cancer, leading to a more pronounced prothrombotic milieu in pancreatic cancer[54]. The involvement of tumor cell-derived TF extends beyond its role in thrombin generation within the cancer milieu; TF also influences its own expression in both malignant and host vascular cells. It promotes tumor progression by upregulating the expression of vascular endothelial growth factor (VEGF). Another distinct procoagulant in tumor cells is CP, which activates factor X independently of FVII. The extent to which CP interacts with TF remains to be fully elucidated[55]. Moreover, CRC tissues express acetylheparinase, which promotes tumor progression by increasing TF expression and interacting with TF pathway inhibitors (TFPI) on the surface of endothelial and tumor cells. This interaction leads to the dissociation of TFPI, thereby augmenting cell surface TF activity[56].

Tumor-derived EVs, which carry lipids, proteins, and nucleic acids, are actively secreted by a diverse array of tumor cells during processes such as activation, apoptosis, or malignant transformation[57]. These EVs have garnered significant interest in the context of prethrombotic disorders due to their increased numbers and thrombogenic activity, highlighting their critical role in thrombosis pathogenesis[58]. Particularly noteworthy in tumor tissues are TF+ EVs. The high concentration of negatively charged phospholipids on the surface of these EVs significantly enhances their TF activity. Studies have established that the procoagulant activity of tumor-derived EVs is primarily due to the presence of TF[59,60]. Elevated levels of TF+ EVs have been identified in CRCs; however, their activity varies from that observed in patients with pancreatic cancer[42]. In addition, TF+ EVs have been shown to lead to the establishment of a thrombotic state in patients with cancer[61-63]. Wang et al[64] showed that only TF-positive tumor-bearing mice had elevated levels of TF+ EVs and enhanced thrombosis in the saphenous vein FeCl₃ injury model. A recent meta-analysis including six studies concluded that TF-containing EVs are associated with an increased risk of VTE in cancer patients[65]. Microsatellite Instability high CRC is characterized by its strong immune response, leading to increased production of inflammatory cytokines and increased release of procoagulant EVs[66]. In addition, microRNAs (miRNAs) contained in EVs have important biological functions, and they can influence gene expression in distal cells. A recent nested case-control study addressed the link between miRNA expression and thrombogenesis in CRC. The authors compared the tumor expression miRNA profiles of patients with CRC with and without VTE. The primary results suggested differential expressions of 19 tumor miRNAs in VTE cases compared with controls, with hsa-miR-3652, hsa-miR-92b-5p, and hsa-miR-10394-5p being the most significantly downregulated in the patients with VTE[67]. The miRNA profile of tumor cells correlated with CAT and may contribute to the hypercoagulability observed in CRC. Another study showed that the generation of EVs is dependent on oncogenes. The authors found that cells overexpressing the MYC and AURKB genes release significantly more EVs than control cells. They also found that an inverse relationship between MYC upregulation and RAS/MEK/ERK signaling pathway activation regulates the release of EVs[68]. The potential use of EVs as predictive biomarkers for VTE risk in patients with cancer remains a subject of ongoing investigation. Current clinical trials are assessing the viability of measuring TF+ EVs as VTE predictors in patients with cancer[69]. Given the evident role of EVs in both thrombosis and cancer progression, exploring ways to modulate their release and activity could hold substantial clinical importance.

Under physiological conditions, plasma coagulation FVII is predominantly synthesized in the liver, primarily by hepatocytes[70]. However, FVII expression extends beyond hepatic production. It can also occur in monocytes within cancerous tissues and under inflammatory disease contexts[71]. This ectopic expression of FVII has been particularly observed in digestive tract tumors[72]. For instance, studies have consistently shown endogenous expression of FVII in colon carcinoma cell lines[73,74]. Ectopic FVII molecules are functional, partly due to the expression of c-glutamyl carboxylase in cancer cells, an enzyme that facilitates the necessary post-translational modifications for proper localization of FVII at the cell membrane[74]. Coagulation factor VIII (FVIII) is another crucial component of the coagulation cascade. Its elevated activity has been linked to increased risks of both primary and recurrent VTEs[75,76]. Patients with CRC have been reported to present high FVIII levels[77]. Results of a retrospective study showed higher FVIII levels in patients with cancer with a history of thrombosis than in matched controls without thrombosis[78]. This finding was corroborated in a prospective cohort study, which demonstrated that highly expressed FVIII plasma levels were a significant risk factor for VTE in patients with cancer[79]. Despite this evidence for the correlation between VTE risk and FVIII levels, the exact impact of malignant tumors on FVIII plasma levels warrants further investigation. Tumor cells seem to influence the host fibrinolytic system, they express fibrinolytic proteins and interact with their inhibitors including plasminogen activator inhibitor-1 (PAI-1) and plasminogen activator inhibitor-2. PAI-1 inhibits fibrinolysis and has been associated with an increased thrombosis risk[80]. In a CRC mouse model, administration of bevacizumab, an anti-VEGF drug, led to increased levels of PAI-1 and thromboses. This effect was alleviated by the use of a PAI-1 inhibitor[81]. Additionally, a study in patients with pancreatic cancer suggested that high PAI-1 antigen and activity levels could predispose to VTE[82].

Tumor cell-induced procoagulant properties of host cells

In CRC, interactions between tumor cells and host cells are mediated by two distinct mechanisms: Direct adhesions and secretion of soluble mediators. These interactions produce significant changes in the host cells, leading to the induction of a procoagulant phenotype. Figure 3 depicts the complexities of these interactions and their impact on host cell behavior.

Figure 3 Induction of procoagulant phenotype in host cells by colorectal cancer cells: A cascade of cellular interactions. This section delineates how colorectal cancer cells orchestrate a procoagulant environment through dynamic interactions with host cells. These interactions occur via direct adhesion or through the release of soluble mediators, leading to the expression of a procoagulant phenotype in host cells. A key mechanism involves the secretion of platelet aggregation agonists by tumor cells, stimulating platelet aggregation. Tumor cells also express Podoplanin, which activates platelets. These activated platelets are pivotal in inducing neutrophil extracellular traps from neutrophils, further promoting thrombus formation by trapping and activating additional platelets. The role of activated platelets extends to enhancing blood hypercoagulability through interactions with endothelium-bound von Willebrand factor. In parallel, cancer cells secrete granulocyte colony-stimulating factor and interleukin-6 (IL-6), which activate neutrophils. The secretion of IL-1β, tumor necrosis factor-alpha, and lipopolysaccharide by tumor cells induces a thrombotic mechanism in monocytes, characterized by increased expression of procoagulant tissue factor. Moreover, tumor-infiltrating macrophages express coagulation factors II, V, VII, and X on their surfaces, contributing further to the procoagulant milieu. Finally, the secretion of inflammatory mediators by tumor cells alters the expression of endothelial cellular products such as thrombomodulin (TM), leading to elevated levels of soluble TM and a concurrent decrease in surface TM expression, thereby diminishing the anticoagulant properties of endothelial cell membranes. MPO: Myeloperoxidase; NETosis: Formation of neutrophil extracellular traps. Created in BioRender (Supplementary material).

Substantial evidence supports the role of platelets in fostering a hypercoagulable state in patients with CRC[83]. Malignant tumors induce platelet adhesion and aggregation either through direct cancer cell-platelet interactions or via tumor cell-secreted platelet aggregation stimulants, including ADP, thrombin, matrix metalloproteinases, and IL-6[21,84,85]. Elevated circulating levels of soluble P-selectin, correlated with higher VTE rates, have been observed in patients with cancer[86]. During the adhesion process, platelets express P-selectin, which binds tumor cells, forming aggregates and facilitating tumor proliferation and metastasis[87]. This interaction accelerates tumor growth and shields tumor cells from immune surveillance. Platelets aggregate around tumor cells, forming a ‘platelet cloak’ that effectively conceals them from natural killer cells[88]. The overproduction of mucins also generates a physical and biochemical barrier that protects tumor cells from recognition and attack by immune cells[89]. Mucin binds to P-selectin on platelets and promotes platelet adhesion, an interaction mechanism that may further consolidate the barrier function of tumor cells and lead to a more efficient immune escape[90]. MUC4 enhances the survival and extravasation of disseminated tumor cells by interacting with platelets[91]. Colorectal tumor cells might also harness Podoplanin, a sialoglycoprotein located on cell membranes, which can activate platelets[92]. Podoplanin, a ligand for the platelet receptor CLEC-2, induces platelet aggregation, a process intimately linked with tumor metastasis and malignancy progression[93]. Moreover, platelets interact with ECs, contributing to CAT. This was evidenced in a deep vein thrombosis mouse model, where platelets were shown to enhance blood hypercoagulability through interaction with endothelial-bound von Willebrand factor (vWF)[94]. Lastly, platelets facilitate thrombosis by activating the coagulation cascade, leading to thrombin generation. This occurs by the exposure of PS on their outer membrane, which serves as a platform for initiating fibrin clot formation[95].

Leukocyte counts are commonly high in patients with CRC, and this increase is associated with a high VTE risk[96]. Leukocytes, particularly neutrophils and monocytes, are significantly influenced by tumor cells, adopting a procoagulant phenotype. These cells are activated by direct contact with tumor cells or via the release of inflammatory cytokines into the bloodstream. CRC cells secrete factors like granulocyte colony-stimulating factor (G-CSF), granulocyte macrophage colony-stimulating factor, and IL-6, elevating the circulating levels[97]. G-CSF increases neutrophil counts and stimulates their activation[17]. Activated neutrophils release various procoagulant enzymes such as elastase, cathepsin G, and myeloperoxidase (MPO)[98].

Activated platelets in patients with CRC have a pivotal function in the development of hypercoagulability, interacting with various host cells. These interactions include stimulating leukocytes to release neutrophil extracellular traps (NETs), which are implicated in venous thromboses[99]. NETs, composed of desmosomal chromatin and proteins like neutrophil elastase (NE), tissue protease G, and MPO, were initially identified as a defense mechanism against pathogens[100-102]. However, in the context of CRC, NETs contribute to a prothrombotic state and thromboses vis multiple pathways[103,104]. NETs provide a structural scaffold for platelet and fibrin adhesion, aiding thrombus formation. Moreover, NETs components such as NE and cathepsin G can activate the coagulation pathway, promoting clotting[105]. Moreover, NETs capture and activate platelets, fostering a procoagulant phenotype and accelerating development of thrombi[106]. Zhang et al[107] provided significant insights into the role of NETs in CRC thromboses. Their study, comparing 60 patients with CRC with 20 healthy controls, revealed a higher prevalence of NETs in patients with cancer, correlating with disease progression.

The role of activated monocytes and macrophages in CAT is well-documented. Macrophages infiltrating tumors adopt a locally activated, procoagulant state, which contributes to the deposition of fibrin within tumor tissues[108]. A study on advanced CRC demonstrated that blood monocytes with a procoagulant phenotype increase the risk of intravascular coagulation and thromboembolic complications[109]. Monocytes distinguish themselves from other circulating blood cells through their capacity to synthesize and express procoagulant TFs upon activation. This activation is triggered by cytokines like IL-1β, TNF-α, and lipopolysaccharide on their surface[110]. These mediators, often secreted by cancer cells, trigger monocyte-mediated thrombotic mechanisms. Additionally, tumor-infiltrating macrophages have been identified as expressing coagulation factors II, V, VII, and X, further implicating them in CAT. Recent research has expanded this understanding by showing that blood monocytes can release extracellular traps (ECTs) in response to various inflammatory stimuli[111]. However, the exact role of these ECTs in the context of CAT, remains unclear.

Under normal physiological conditions, ECs maintain blood flow by providing a smooth, antithrombotic surface that prevents platelet adhesion and coagulation. In patients with CRC, several factors disrupt this equilibrium. Tumor cells can activate ECs either through direct cell-to-cell contact, as observed in CRC cases, or by releasing inflammatory mediators and acute-phase proteins that stimulate endothelial activation[112,113]. Cytokines like IL-1b and TNF-α are key regulators altering ECs functions related to hemostasis, by altering the expression of thrombomodulin (TM), TF, vWF, selectins, and fibrinolytic proteins[114]. TM, an endothelial membrane receptor with significant anticoagulant properties, forms a complex with thrombin to activate protein C, a natural anticoagulant[115]. However, in patients with CRC, an increase in soluble TM levels and a decrease in surface TM expression mark the loss of the endothelial anticoagulant function[116]. Under inflammation, TNF-α from ECs stimulates the expression and release of soluble TF, a potent prothrombotic agent. The shift from an anticoagulant to a prothrombotic endothelium in CRC is marked by the upregulated procoagulant TF and the downregulated TM/protein C system[117]. In addition, activated ECs release soluble adhesion molecules like E-selectin and P-selectin, which have been found at elevated levels in patients with CRC with thrombosis[118]. These molecules, particularly P-selectin, enhance VTE by recruiting leukocytes and promoting platelet adhesion and aggregation, thereby increasing the thrombotic risk[119]. Moreover, endothelial-derived nitric oxide, an inhibitor of platelet adhesion and aggregation, is reduced in CRC, further contributing to the thrombotic milieu[120]. Lastly, while the counts of circulating ECs are elevated during endothelial injury in solid tumors, their specific role in CAT remains unclear[121].

CRC GENE PERSPECTIVE

Molecular biology research has increasingly clarified the function of oncogenes in tumor transformation and their impact on the expression of coagulation-related proteins within cancerous tissues. These processes occur after mutations and deletions in genes like KRAS, EGFR, MET, PTEN, and TP53. Historically, hypercoagulable states and thrombosis in patients with cancer were perceived as nonspecific byproducts of cancer progression and associated vascular disruptions, often attributed to vascular hyperpermeability and inflammation[13]. However, emerging evidence now points to these coagulopathies being cancer-specific phenomena[13]. Rak and colleagues have argued that the coagulation system’s functionality is altered by different types of cancer cell, noting significant variations in VTE risk across different cancers. For instance, CRCs display a higher VTE risk than skin, breast, and prostate cancers[29,122]. This suggests that specific tumor genotypes might directly influence the coagulation system or induce changes in the tumor microenvironment[123]. Additionally, these pathways can directly dysregulate coagulation effectors via different mechanisms: The aberrant expression of TFs, the induction of ectopic coagulation genes, or the release of EVs containing TF-positive genes into the systemic circulation[124,125].

Yu et al[47] have produced evidence showing the role of TF expression in human CRC cells. Their results delineate how TF expression is regulated by two pivotal translational events that mark the progression of this disease: Activation of the KRAS oncogene and inactivation of the p53 oncogene. Intriguingly, this regulation is contingent upon the activity of the MEK/mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) pathways. These findings implicate the MAPK and PI3K pathways as potential participants in the upregulation of TF in CRC cells, which lead to a high VTE risk. Further corroborating this, Regina et al[126] validated these observations in actual CRC samples. Beyond the scope of KRAS, mutations in PTEN are also implicated in the increased expression of TF, adding another layer of complexity to this regulatory mechanism[127]. Additionally, less expected genetic variants have been explored to explain the development of VTE among patients with cancer. A notable study revealed that patients with CRC with the β3-integrin rs3809865 A/A genotype faced a heightened risk of VTE than those with A/T or T/T genotypes[128]. KRAS mutations are present in approximately 40% of patients with CRC, usually occurring at codons 12, 13, and 61, resulting in constitutive activation of the RAS/MAPK and PI3K/AKT signaling pathways. Experimental studies have shown that mutant KRAS proteins promote EC activation and upregulate TF expression through multiple mechanisms. Specifically, oncogenic KRAS induces activation of the RAF-MEK-ERK and PI3K-AKT pathways, leading to enhanced transcription of TF through AP-1 and NF-kB signaling (Figure 4)[129,130]. In addition, KRAS mutations increase the secretion of VEGF, stimulating endothelial TF expression and promoting the release of procoagulant EVs from tumor cells, thereby amplifying the coagulation cascade response[131]. Clinical evidence supports this mechanism as patients with metastatic CRC harboring KRAS mutations have a significantly increased risk of VTE (OR, 2.21; 95%CI: 1.08-4.53), reinforcing the role of KRAS-driven hypercoagulability[132]. Although patients with BRAF mutations in CRC have also exhibited high rates of VTE, the number of cases has been insufficient to definitively establish a role for BRAF as a risk factor for VTE[133].

Figure 4 The mechanism diagram illustrates that KRAS gene mutations lead to an increase in tissue factor expression through constitutive activation of the RAS/mitogen-activated protein kinase and phosphatidylinositol 3-kinase/AKT signaling pathways. Created in BioRender (Supplementary material).

The coagulation phenotype of cancer cells is increasingly recognized as a key mechanism bridging the genetic evolution of the disease with its biological and clinical characteristics[8]. Research has shed light on the heightened incidence of CRC in individuals predisposed to certain thrombotic conditions, particularly those with a factor V Leiden mutation[134]. This emerging evidence proposes a change in basic assumptions: Procoagulant events in CRC may not merely be concurrent phenomena, they may be active contributors to tumor growth and progression, and potentially, initiators of malignant transformation. From a therapeutic standpoint, this revelation underscores the potential of targeting coagulation disturbances for cancer management. The refinement of strategies to control coagulation perturbations could pave the way for novel approaches in the treatment, control, and prevention of cancer. Such strategies may mitigate the direct impacts of cancer and also impede its progression by addressing underlying coagulation-related mechanisms.

The discovery that coagulation events can be genetically driven is important. The correlation of KRAS mutations in colorectal and lung cancers with heightened VTE risks stands in contrast to the IDH1/2 mutations in glioblastomas, which appear to reduce the VTE risk[132,135,136]. Given the labor-intensive work of examining each gene’s role in CAT, a more efficient approach might involve an unbiased genomic exploration. This strategy could unveil novel biomarkers and potentially reveal new therapeutic targets for managing CAT. Moreover, a comprehensive understanding of the primary biological mechanisms that precipitate hypercoagulability and consequent VTE necessitates an in-depth evaluation of the genetic landscapes specific to each cancer subtype. This approach is crucial, as the genetic determinants associated with CAT are likely to vary according to the unique processes inherent to each cancer type. Such targeted investigations promise to identify patient subgroups with high VTE risks, enabling the development of precise and effective anticoagulation strategies. This, in turn, may significantly improve thrombosis prevention in patients with cancer, marking a critical step forward in managing this complex and challenging clinical issue.

CONCLUSION

In this article, we systematically reviewed the molecular and clinical mechanisms of CAT in patients with CRC, focusing on the procoagulant properties of tumor cells themselves and their interactions with host cells. In the realm of CRC, the challenge posed by CAT is substantial, given its significant contribution to increased morbidity and mortality. Despite years of dedicated research aimed at unraveling the mechanisms underlying CAT and identifying high-risk patient profiles, advancements have been gradual. Significant progress in the treatment of CAT in CRC will be made with the development of personalized medical approaches that integrate newly discovered biomarkers and targeted therapies. Emerging biomarkers such as CTCs and tumor-derived EVs are expected to enable precise risk stratification and monitoring. In terms of prevention strategies, regular monitoring of coagulation function and the use of thrombosis risk assessment tools can help clinicians implement targeted preventive anticoagulant therapy. On the therapeutic side, the development and clinical integration of novel oral anticoagulants, such as rivaroxaban and edoxaban, are expected to significantly improve the therapeutic efficacy of conventional anticoagulants, such as low molecular weight heparin and warfarin. Studies focusing on the tumor microenvironment have revealed interactions between cancer cells and the coagulation system. TF expression is frequently upregulated in CRC and may be a new therapeutic target. In addition, the integration of artificial intelligence and machine learning with large-scale clinical and genomic data is expected to revolutionize predictive models for CAT. Future studies will also explore the role of immunotherapy and anti-inflammatory treatments and their interactions on cancer progression with thrombotic events. These advances will pave the way for a comprehensive multidisciplinary approach addressing CAT and improving the quality of life and survival of patients with CRC.