**4. The use of biomarkers for risk assessment for VTE in cancer patients**

Despite the well documented association of cancer with increased risk of thrombosis, clinical studies have not consistently demonstrated improved outcomes with

Venous Thromboembolism in Cancer Patients 89

e. **The prothrombin fragment** 1+2 (F1+2) is released when activated factor X cleaves prothrombin to thrombin and reflects the in-vivo thrombin generation. A systematic activation of coagulation has been observed in cancer patients which is reflected by elevated plasma levels of global coagulation markers, such as D-Dimer or prothrombin

f. **Tissue factor (TF)** is a transmembrane glycoprotein present on subendothelial tissue, platelets, and leukocytes that initiates coagulation and plays a critical role in regulating hemostasis and thrombosis (174-175). TF expression has been shown to be associated with increased angiogenesis in various solid neoplasms, including hepatocellular, colorectal and prostate cancers as well as in haematologic malignancies and play a role in cancer-associated thrombosis (176-177). TF induction was shown to be an early event in the development of pancreatic cancer and that the level of TF expression correlates with increased angiogenesis and with subsequent development of symptomatic VTE

VTE was 4-fold more common (p = 0.04) among patients with high TF-expressing carcinomas (20%) than among those with low TF-expressing carcinomas (5.5%). There is a potential for circulating TF to be used as predictive biomarker for pancreatic cancer associated VTE. VTE was 4-fold more common (p = 0.04) among patients with high TFexpressing carcinomas (26.5%) than among those with low TF-expressing carcinomas (5.5%). In another study, TF expression correlated with subsequent VTE in a series of patients with

Furthermore, a retrospective analysis of cancer patients without VTE, revealed a 1-year cumulative incidence of VTE of 34.8% in patients with TF-bearing MPs versus 0% in those

g. **Soluble P-selectin (sPS):** This is a cell adhesion molecule found in the membranes of platelets and endothelial cells (Weibel–Palade bodies) where it can function as a receptor and mediate cell adhesion via binding to several ligands. The interaction of sPS with PSGL-1 expressed on the majority of leukocytes results in the release of procoagulant, tissue factor-rich microparticles (MPs) from leukocytes, endothelial cells, platelets and cancer cells (179). In case-control studies of non-cancer patients with a history of VTE and healthy subjects without a history of venous or arterial thrombosis, high plasma levels of sPS have been demonstrated to be strongly associated with VTE (27, 35). In a multivariate analysis of the prospective observational Vienna Cancer and Thrombosis Study, elevated sPS (cutoff level, 53.1 mg/mL) was a statistically significant risk factor for VTE after adjustment for age, sex, surgery, chemotherapy and radiotherapy (HR = 2.6) and the cumulative probability of VTE after 6 months was 11.9% in patients with high sPS and 3.7% in

C-reactive protein (CRP) is an inflammatory marker produced by the liver and adipocytes. In a prospective study, CRP was significantly associated with increased risk of VTE by

was 14% in patients with elevated FVIII:C levels

fragment 1+2 (F1+2) (49, 173).

(176-179).

ovarian cancer (173).

**C- Reactive Protein** 

multivariate analysis (181-183)

without detectable TF-bearing MPs (p= 0.002).

those normal levels (p = .002) (180).

almost 3-fold increased VTE risk **(172).** Cumulative probability of VTE after 6 months

thromboprophylaxis in all groups of cancer patients, and hence their risk for VTE in view of the heterogenity of cancer (166).

Moreover, treatment of VTE in patients with cancer or use of pharmacological agents for thromboprophylaxis is more difficult and is associated with considerable therapeutic challenges in view of thrombocytopenia caused by some chemotherapeutic agents and morbidity associated with VTE in often medically compromised cancer patients (32, 167)

Therefore, identification of a high-risk subgroup of cancer patients who will benefit from primary thromboprophylaxis is well justified. Recent data have identified multiple clinical risk factors as depicted under patient-, disease- and treatment-related risk factors above as well as biomarkers predictive of VTE in cancer patients. Biomarkers associated with increased risk of cancer associated VTE include leukocyte count, platelet count, and levels of tissue factor, P-selectin, D-dimer and CRP as discussed below.


Pathophysiology and Clinical Aspects of 88 Venous Thromboembolism in Neonates, Renal Disease and Cancer Patients

thromboprophylaxis in all groups of cancer patients, and hence their risk for VTE in view of

Moreover, treatment of VTE in patients with cancer or use of pharmacological agents for thromboprophylaxis is more difficult and is associated with considerable therapeutic challenges in view of thrombocytopenia caused by some chemotherapeutic agents and morbidity associated with VTE in often medically compromised cancer patients (32, 167)

Therefore, identification of a high-risk subgroup of cancer patients who will benefit from primary thromboprophylaxis is well justified. Recent data have identified multiple clinical risk factors as depicted under patient-, disease- and treatment-related risk factors above as well as biomarkers predictive of VTE in cancer patients. Biomarkers associated with increased risk of cancer associated VTE include leukocyte count, platelet count, and levels of

a. **Leukocyte count:** Leukocytosis was identified as independent risk factor associated with increased risk of VTE in cancer patients before initiating chemotherapy (OR 2.0). VTE occurred in 4.5% patients with baseline leukocytosis, WBC ≥11x109/L, compared to (1.8%) without leukocytosis (p < 0.0001). In a prospective observational study of 3303 ambulatory cancer patients: "Awareness of Neutropenia in Chemotherapy" Study Group Registry, leukocyte count >11.0x109/L was also reported to be independently associated with an increased risk of subsequent VTE. Leukocytosis may be a marker of the aggressiveness of the cancer cells or represent a direct causative role in mediating

cancer-associated thrombosis, through, as yet, unknown mechanisms (168-169). b. **Platelet count:** Thrombocytosis is often observed in cancer patients and elevated platelet counts correlates with an activation of coagulation. In several studies of cancer patients, an elevated platelet count (≥350x109/l) prior to starting chemotherapy was found to be strongly associated with VTE (21, 19). The incidence of VTE was 4-7.9% in patients with a pre-chemotherapy platelet count ≥350x109/l compared to 1.25% in patients with lower platelet counts. The increased risk of VTE with higher platelet counts persisted while the patients were on chemotherapy and these patients had a 3-

c. **D-Dimer** is a degradation product of cross-linked fibrin that is formed after thrombingenerated fibrin clots have been degraded by plasmin. Elevated fibrin D-dimer level (HR, 1.8) and elevated prothrombin split products (HR, 2.0) have recently been shown to be associated with increased risk of VTE in a large prospective study of cancer patients (170). D-dimer was also elevated in metastatic breast cancer patients compared to normal controls (171). These and other data suggest that D-dimer levels may be a

d. **Clotting factor VIII (VIII:C)** This factor plays an important role in the coagulation cascade. In non-cancer patients, a high FVIII: C activity has been established as a risk factor for primary and recurrent VTE (25-26, 51). In a prospective cohort study, a significant association was found between FVIII:C levels and the risk of symptomatic VTE in cancer patients (51). In an analysis of cancer patients including solid cancers and haematological malignancies, the cumulative probability of VTE after 6 months was 14% in patients with elevated FVIII (cut-off: 232%) as opposed to 4% in those with normal levels (p = 0.001). These results demonstrate that elevated FVIII:C levels in cancer patients proved to be a valuable, independent risk marker for VTE, predicting an

tissue factor, P-selectin, D-dimer and CRP as discussed below.

the heterogenity of cancer (166).

fold higher rate of VTE (32).

predictor of VTE in cancer patients.

almost 3-fold increased VTE risk **(172).** Cumulative probability of VTE after 6 months was 14% in patients with elevated FVIII:C levels


VTE was 4-fold more common (p = 0.04) among patients with high TF-expressing carcinomas (20%) than among those with low TF-expressing carcinomas (5.5%). There is a potential for circulating TF to be used as predictive biomarker for pancreatic cancer associated VTE. VTE was 4-fold more common (p = 0.04) among patients with high TFexpressing carcinomas (26.5%) than among those with low TF-expressing carcinomas (5.5%). In another study, TF expression correlated with subsequent VTE in a series of patients with ovarian cancer (173).

Furthermore, a retrospective analysis of cancer patients without VTE, revealed a 1-year cumulative incidence of VTE of 34.8% in patients with TF-bearing MPs versus 0% in those without detectable TF-bearing MPs (p= 0.002).

g. **Soluble P-selectin (sPS):** This is a cell adhesion molecule found in the membranes of platelets and endothelial cells (Weibel–Palade bodies) where it can function as a receptor and mediate cell adhesion via binding to several ligands. The interaction of sPS with PSGL-1 expressed on the majority of leukocytes results in the release of procoagulant, tissue factor-rich microparticles (MPs) from leukocytes, endothelial cells, platelets and cancer cells (179). In case-control studies of non-cancer patients with a history of VTE and healthy subjects without a history of venous or arterial thrombosis, high plasma levels of sPS have been demonstrated to be strongly associated with VTE (27, 35). In a multivariate analysis of the prospective observational Vienna Cancer and Thrombosis Study, elevated sPS (cutoff level, 53.1 mg/mL) was a statistically significant risk factor for VTE after adjustment for age, sex, surgery, chemotherapy and radiotherapy (HR = 2.6) and the cumulative probability of VTE after 6 months was 11.9% in patients with high sPS and 3.7% in those normal levels (p = .002) (180).

#### **C- Reactive Protein**

C-reactive protein (CRP) is an inflammatory marker produced by the liver and adipocytes. In a prospective study, CRP was significantly associated with increased risk of VTE by multivariate analysis (181-183)

Venous Thromboembolism in Cancer Patients 91

properties of the tumor cells can influence the hypercoagulable state of patients with these malignancies by several mechanisms. Of interest, oncogenes responsible for haematological neoplastic transformation in leukemia also may be involved in haemostatic activation.

Patients with brain tumors are at particularly high risk for VTE, and many studies found that the hazard for deep vein thrombosis in patients with malignant glioma may reach 28% (187-188) This high risk is maintained throughout the course of an active disease and during treatment, and not just in the immediate postoperative period (188). Risk factors for VTE in patients with glioma include the presence of paraparesis, a histologic diagnosis of glioblastoma multiforme, age ≥ 60 years, large tumor size, the use of chemotherapy, and length of surgery of > 4 hours. Because of the high incidence of VTE, patients who are treated for brain tumors are usually considered for long term prophylactic anticoagulation

The incidence of clinical VTE in patients with malignant lymphoma reportedly ranges between 6.6% and 13.3% (187, 190-194) In one study, 50% of patients had a bulky tumor compressing a vein, 25% of patients had a central catheter at the thrombosed vein, and, in the other patients, thrombosis was attributed to paraneoplasia or to chemotherapy (187)

Conlon SJ et al (194) reported the results of retrospective analysis of patients with a total of 18 653 cases on the NCI Working Formulation: there were 5496 low grade NHL, 12 251 aggressive NHL and 906 high-grade lymphoma cases. The cumulative incidence (CI) of VTE 24 months from diagnosis was 4.0%. The CI of VTE at 24 months were significantly different for the distinct lymphoma groups (p< 0.001, Chi-square and comparisons showed this difference to be significant only between the low grade and other histologies. Of 742 cases that had VTE, 454 died within 2 years (61%). For those without VTE, 7274 of 17911 (41%) died within 2 years. This difference was statistically significant (p < 0.001, Chi-square).

A population- based cohort study (1993-1999) to determine the incidence and risk factors associated with development of VTE among Californians diagnosed with acute leukemia (1993-1999) was reported in Blood 2009, by Ku GH et al (195). Among 5394 cases of AML, the 2-year cumulative incidence of VTE was 281 (5.2%) and 64% of VTE events occurred within 3 months of AML diagnosis. The authors reported that, in AML patients, female sex, older age, number of co-morbid conditions, presence of CVC were significant predictors for development of VTE within one year following diagnosis of acute leukemia but the event of VTE was not associated with poor survival in AML patients in the studied group. Among 2482 case of ALL, the 2-year incidence of VTE was 4.5% and risk factors in this group were presence of CVC, older age and number of chronic co-morbidities. In this study, development of VTE in ALL patients was associated with a 40% increase of dying within

In the abstract #6595 published in JCO 2011 by Luong NV et al from MD Anderson Cancer Center, USA (196) of a retrospective chart review to determine the prevalence of VTE prior

• **VTE in Central Nervous System Lymphoma** 

as deemed appropriate for a particular patient (189-190).

• **VTE in Non-Hodgkin Lymphoma** 

• **VTE In Patient with Acute Leukemia** 

one year.


Table 1. Summary of the Risk Factors for VTE in Cancer Patients
