The primary serum constituents resulting from platelet-initiated coagulation, growth, and wound repair are derived from intracellular sources (43–47). Circulating, resting platelets contain multiple storage granules: 1) alpha-granules (α-granules) 2) dense granules (δ-granules) 3) lysosomes, and 4) microparticles (48–50). There is also a newly described T-granule that primarily contains toll-like receptors 9 (51).

Of the various types of platelet granules, the most common are α-granules, which constitute 10% of the platelet volume and number 50-80 per platelet (52). Two pathways contribute to α-granule formation. De novo synthesis occurs when the trans-Golgi network generates a core structure that attracts clathrin to form a lattice structure and interacts with coat proteins like adaptor-protein 1 resulting in clathrin-coated pits. After invagination, these pits bud forming membrane-bound vesicles that end up in early endosomes. A strictly endocytic vesicle pathway also exists involving adaptor-protein 2 that is involved in forming subsets of endosomes. Alpha-granule subset lineages mature in multivesicular bodies through the engagement of vacuolar protein sorting-associated protein (-33B and -16B) that are involved in sorting certain cargo from the trans-Golgi network. Neurobeachin-like protein 2 is another molecule that helps drive α-granule development and secretion. Mature α-granules are released via exocytosis following platelet stimulation via a process involving dynamin-related protein-1 and other cytoskeletal elements (53–55).

Platelets release significant amounts of factors from α-granules such as α2-macroglobulin, plasminogen activator, plasminogen, and plasminogen-activator inhibitor type-1 that help with fibrinolysis, clot dissolution, inflammation resolution, and wound repair (113–117). They also release tissue remodeling enzymes involved in wound repair that include matrix metalloproteinases (MMP-1, MMP-2, MMP-3, MMP-9, MT1-MMP, MMP-14), tissue inhibitor of metalloproteinases (TIMPs; TIMP-1, TIMP-2 and TIMP-4), as well as a disintegrin and metalloproteinases (ADAMs; ADAM-10, ADAM-17, ADAMTS-13) (118, 119). Platelet-activating factor induces the expression of a number of tissue-remodeling proteases (120–122). PDGF also stimulates tissue remodeling hepatic stellate cell (HSC) proliferation and fibrosis in the liver that involves tissue-remodeling proteases (123). MMPs secreted by activated platelets play an integral role in tumor spread and the metastatic cascade by contributing in tissue remodeling and stimulating tumor-cell transmigration and invasion of surrounding tissues, blood vessels, and liver sinusoids (124–128).

Lysosome function in platelet biology and hemostasis is not well characterized (50, 55, 90, 91, 98, 129–134). Because they release phospholipase A, protease, and glycohydrolase enzymes, this suggests that lysosomes play a role in platelet responses and dissolution of clots (129, 132–134). Successive waves of platelet activation, adhesion, aggregation, and stabilization allow these first responders to activate specific subsets of cytoskeletal changes to recognize and secure vascular lesions.

The body is thought to produce and clear 10¹¹ platelets per day (135, 136); clearance occurs primarily in the liver and spleen. Clearance mechanisms include senescence and apoptosis driven by Bcl-xL and the proapoptotic molecules Bax and Bak, which set the internal clock for platelet lifespan in conjunction with BH3-only proteins, mitochondrial permeabilization, phosphatidylserine exposure, and lectin-mediated recognition of platelet glycans (135–137). The vWF or antibodies binding to platelet surface glycoproteins under fluid shear induce mechanosensory signaling that leads to ADAM17 and phosphatidylserine exposure along with desialylation (138). These molecular changes are also commonly associated with platelet aging-related clearance and thrombocytopenia found in type 2B von Willebrand disease (139). Recognition of platelet glycans by the Ashwell-Morell receptor leads to clearance of senescent platelets by hepatocytes, macrophages, and other resident liver- or spleen immune cells. The complexities and temporal state of the platelet life-span clock and surface recognition by resident liver-immune cells is likely to influence the dynamics of liver diseases by selectively eliminating aging platelets and allowing fully functional platelets to exert their normal biological function.

Elevations in circulating platelet-to-lymphocyte ratios are often associated with inflammation and poor outcomes along with being linked to infections, inflammatory diseases, liver disease and cancer (140–146). The predictive value indicative of poor outcomes linked to elevated platelet-to-lymphocyte ratios has shown clear associations with liver transplantation, particularly when performed for hepatocellular carcinoma and liver metastasis (147–152).

Evidence exists in support of the predictive value of platelet-to-lymphocyte ratios as a general biomarker of inflammation but prospective case-control studies in each disease will help to further establish this hypothesis.

Ultrastructural studies of experimental metastasis to the liver were first reported by Dingemans (153, 154) ( Figure 5B ), who injected mammary carcinoma cells into syngeneic C57/Bl6 mice. Almost immediately after mesenteric-vein injection, tumor cells had formed large emboli in the portal branches. Tumor emboli adhered to the vascular wall without completely occluding the lumen. On the side of the embolus facing the vascular lumen, aggregated platelet clusters were found as part of the first response in the liver. On the luminal side of the platelet clusters, leukocytes, especially neutrophils, had adhered. Erythrocytes, by contrast, were present near the non-platelet-involved tumor cell surfaces, potentially suggesting that immune cells were preferentially attracted to the aggregated platelets. An outer zone consisting of the platelet aggregates was formed by degranulated- and more or less spherical platelets. The platelet aggregate centers consisted of closely packed, elongated elements along with small amounts of fibrin within heterogenous emboli, which were gradually displaced during metastatic growth and disease progression.

Along with regulating platelet number, the liver also plays an important role in coagulation. Both coagulation and anti-coagulant proteins are primarily made in the liver; thus, any liver disease can potentially dysregulate coagulation (156, 194–196). Most coagulation factors are synthesized by the parenchymal cells of the liver (factors I, -II, -V, -VII, -IX, -X; proteins C, -S, and –Z; fibrinogen; antithrombin; α2-PI; and plasminogen); while factor VIII is produced by liver LSECs (197). Of these, the synthesis of factors II, -VII, -IX, -X; and proteins C and -S are dependent on vitamin K—an important cofactor for regulating coagulation. As the primary storage site for vitamin K, the liver provides conversion of synthetic vitamin K to its active form and produces the bile salts that aid with the absorption of food-based vitamin K (156, 194–196, 198). The liver also plays a role in the clearance of the coagulation products from the bloodstream and regulates anticoagulation by removing activated clotting- and fibrinolytic factors via the hepatic reticuloendothelial system.

Liver disease can manifest through several mechanisms. Overuse of certain drugs or alcohol, metabolic syndrome, diabetes, chronic viral infection, and exposure to toxins are some of the many contributors to liver injury and disease. Depending on the degree of liver damage, individuals with liver disease have deficiencies in clotting enzymes, which manifest as prolonged clotting times in in vitro assays. Chronic liver-disease-associated coagulation disorders result from the inability of the liver to produce or clear clots (194, 199). Imbalances in the synthesis or clearance of clotting factors can increase the risk of bleeding; increasing evidence suggests that these also increase the risk of prothrombotic events (156). The development of coagulopathies is associated with chronic liver disease; circulating levels of some coagulation factors such as vWF and factors II, -V, and -VII have been shown to correlate with the severity of liver disease. Blood coagulation-protein levels can also reflect liver-cell functionality (200–202). In some instances, irregularities in these levels can contribute to the process of liver damage.

Trousseau first reported excessive blood coagulation in cancer patients with elevated platelet counts or thrombocytosis in 1865 (252–254). Since then, numerous studies have reported on thrombocytosis in cancer. In ovarian cancer, thrombocytosis is linked to elevated tumor interleukin-6 and liver-generated thrombopoietin and is associated with shortened overall survival of patients (255). In one study, orthotopic ovarian-mouse models revealed tumor-derived, interleukin-6-stimulated, hepatic-thrombopoietin (TPO) synthesis and paraneoplastic induction of thrombocytosis (255). Liver metastasis along with Fusobacterium in the liver can also trigger the thrombotic Trousseau’s syndrome (256–258). Observed abnormalities associated with coagulation factors like fibrinogen and prothrombin in liver disease could activate platelets (259). In certain conditions, platelets interact and bind with endothelial cells, hepatocytes, hematopoietic stem cells, and Kupffer cells in the liver (173). Upon injury to the liver endothelium, platelets are sequestered within the sinusoid and lead to endothelial-cell activation. This activation releases cytokines that facilitate the infiltration of immune cells (175). Recruitment of neutrophils and macrophages by platelets is of importance in liver disease as they regulate inflammation, fibrinogenesis, and fibrinolysis, and contribute to fibrosis (260). Platelet-mediated neutrophil recruitment and the interaction of platelets with the vasculature have been suggested as mechanisms driving cholestasis-induced liver damage (261). In NAFLD, the severity of inflammation and fibrosis in the liver was shown to directly correlate to the increased platelet turnover (262). In cirrhosis, the elevated levels of vWF promote platelet binding to collagen (263). Damage to the liver can directly affect liver-produced TPO levels. Circulating TPO levels have been shown to negatively correlate with the stage of liver disease, while reduced levels can lead to decreased platelet production or thrombocytopenia. In addition, thrombocytopenia in chronic liver disease is also caused by hypersplenism and increased sequestration of splenic platelets and increased platelet destruction due to aggregation in the liver (264). Increased thrombosis in these patients may consume platelets leading to lower levels in circulation (265). PDGF‐β, a mediator of hepatic fibrosis, is produced by platelets and released upon their activation (266). With the established association between platelet activation and liver fibrosis, anti-platelet drug use could prove beneficial in combating platelet-mediated liver disease and progression to cirrhosis. Recently, a systemic review and meta-analysis report of 3,141 patients concluded that the use of anti-platelet agents (such as aspirin or clopidogrel) was associated with a 32% decreased odds of hepatic fibrosis (adjusted pooled OR 0.68; CI 0.56–0.82, p ≤ 0.0001) (267). However, a prospective cohort study of patients at high risk of cardiovascular events revealed an inverse association between the use of anti-platelet agents and the presence and degree of liver fibrosis (268). A recent study of patients with hepatocellular carcinoma demonstrated that anti-platelet therapy was associated with improvement in overall survival and reduction in liver-related deaths. This study also reported that the use of anti-platelet therapy tended to delay the deterioration of liver function in these patients (269). Recent summaries of outcomes data in cancer prevention have also highlighted the cancer mitigating role of aspirin and other agents (270).

The liver is a common site for cancer metastasis from primary tumors originating in the gastrointestinal tract. Traditionally the preponderance of liver metastasis was felt to be a function of drainage from the intestines into the hepatic circulation; however, the liver may instead represent a favorable microenvironment for metastasis. The liver’s unique regeneration may likewise contribute to creating a receptive metastasis-formation microenvironment involving platelets and the chronic activation of pathways related to the Myc family of regulator genes (287–292). More than half of CRC patients develop liver metastases (associated with poor prognosis) (85, 225) and an increased risk of thrombosis (241, 293). A positive correlation has been observed between high platelet counts and CRC tumor invasiveness and -metastasis, and a negative correlation between high platelet counts and survival in CRC patients (294–296). Platelet hyperactivation measured by elevated platelet mean volume or aggregation (297–301) can be a predictor of cancer progression in CRC (302–304). Genes implicated in CRC are also involved in platelet activation and coagulation suggesting a prothrombotic environment in CRC (305). During cancer progression and metastasis, platelet responses can be modulated by the tumor cells. Circulating tumor cells (CTCs) can stimulate heterotypic tumor-cell-induced platelet aggregation (TCIPA) (29, 85, 306). Tumor cells may get trapped within a TCIPA, go undetected by immune surveillance, be protected from shear forces while in circulation, and have greater capacity to migrate and metastasize (61, 271). Recently, a microfluidic approach was developed to isolate CTCs by targeting platelets that satellite on the tumor-cell surfaces. A significant number of platelet-coated CTCs was observed in metastatic cancer patients of both epithelial (lung and breast)- and non-epithelial (melanoma) tumor origins. Also observed were single cells and clusters, along with CTCs associated with leukocytes (307). Isolating CTCs using platelet markers emphasizes the potential of platelet-coated CTCs that go unnoticed by conventional isolation methods and their possible significance in metastasis.

Tumor cells also release thrombin which serves to activate platelets and results in the formation of tumor thrombi, a form of TCIPA. These thrombi could provide a supportive environment for CTCs by becoming tethered to blood vessels within distant tissues. Given the heightened coagulatory imbalances in the liver driven by cancer or pre-existing- and/or chemotherapy-induced chronic liver disease, these thrombi can become lodged within the liver sinusoids, potentially leading to the initiation of metastasis. Many studies have used anti-platelet drugs with chemotherapy to reduce platelet-mediated tumor-cell survival and metastasis. Platelets were also shown to promote angiogenesis by releasing angiogenic growth factors such as VEGF (308). In a study that provided aspirin along with tamoxifen therapy to patients with breast cancer investigated the impact of aspirin therapy on circulating levels of the proangiogenic-protein VEGF, the antiangiogenic-protein TSP‐1, and platelet-mediated angiogenic-protein release. The aspirin therapy not only impacted angiogenic protein levels but it may modify the angiogenic balance in women treated with tamoxifen therapy. The increase in antiangiogenic TSP‐1 levels without a concurrent increase in pro‐angiogenic VEGF levels suggests an anti-angiogenic balance from aspirin therapy. This study also found less release of platelet angiogenic proteins (309).

In addition to the pathways discussed previously, dysregulated modulation of TGFβ activity can also provide a favorable environment for tumorigenesis (76). Platelets are the major source of latent TGFβ which is released during platelet activation. This process also releases a furin-like proprotein convertase from platelets which in turn activates TGFβ. This enzyme-mediated activation was shown to continue in the damaged area even after platelet-derived TGFβ activation was complete (75). Aberrant expression of TGFβ could lead to liver fibrosis. Elevated levels of vWF can increase platelet binding to collagen in the cirrhotic condition (263). Thus, conditions like liver cirrhosis can increase the activation and aggregation of platelets, with TGFβ and other growth-factor release not only causing platelet binding with collagen, but also providing a tumorigenic niche within these platelet-collagen traps to capture CTCs and promote their growth within the liver.

Minimal residual disease (MRD)—a collection of viable cancer cells that are undetectable by standard imaging methods—differs by tumor type and requires disease-specific stratification based on distinctive organ-related microenvironmental disease characteristics (310). Recent discoveries in the development and use of liquid biopsies based on analyses of biomarkers in body fluids—including blood, urine, and cerebrospinal fluid—are showing potential for use in stratifying MRD. Liquid biopsy analyses can include (1) tumor-derived DNA, -RNA, -miRNA; epigenetic alterations; and proteins present in cell-free plasma or (2) contained in CTCs or (3) circulating exosomes and microvesicles or tumor educated platelets (271, 285, 286, 311–314). To fully understand MRD from a liquid biopsy standpoint, the inclusion of proteomics and metabolomics may be desirable to form a complete integrated molecular profile. To help with this, animal models of MRD involving gut related vascular modes of injection are materializing, which can help represent MRD in CRC (315, 316). Exposing platelets to tissue-factor-expressing tumor cells in the MRD microenvironment, may also trigger the activation cascade that promotes disease progression.

Based on our experience in developing in-house assays for solid tumor ctDNA (317), the detection of blood-based circulating-tumor DNA (ctDNA), a prognostic biomarker highly sensitive for CRC recurrence following curative-intent therapy, identifies MRD that will inevitably develop into clinically detectable, local recurrent- and/or distant metastases (318–321). Currently, ctDNA assays that detect MRD are also commercially available as a standard-of-care tool that enables oncologists to monitor for CRC recurrence. However, when ctDNA is detected, no prospective data directly linked to proven therapies are available to guide clinical management of CRC (322). A clinical need exists to better understand the biology of micrometastases and the role of platelets, especially in microscopic disease. If platelet-based biomarkers (e.g., platelet-to-lymphocyte ratios or micrometastasis-educated platelets) can portray ctDNA-defined MRD before the onslaught of macroscopic, clinically evident metastatic disease, our diagnosis and window for treating microscopic disease will likely improve before the tumor growth outpaces the immune-system ability to eliminate the disease. Presently, little is known from patient samples about the impact of platelets on the biology of micrometastases, while few therapeutics that consider this potential platelet involvement have been tested in patients.