Heparin and the survival benefit in cancer patients: the birth of the hypothesis

Cancer may activate the coagulation system, which has been known since 1823 when Bouillaud described three patients with cancer and venous thrombosis. In addition, cancer cells use this system specifically for their growth and metastasis. The first clinical observation of anticoagulants having an inhibitory effect on cancer growth was reported in a study published in 1992. This trial compared the efficacy and safety of two types of heparin, low molecular weight heparin (LMWH) and unfractionated heparin (UFH), in the initial treatment of patients with venous thromboembolism (VTE). Unexpectedly, the subgroup of patients with VTE who also had cancer at entry had a better survival after 3 months when treated with LMWH for 5 to 10 days compared with those treated with UFH.

A meta-analysis published in 1998 of all VTE treatment studies confirmed this finding. In patients with VTE who had cancer, the observed 3-month mortality rate was 15.0% (46/306) for those initially treated with LMWH compared with 22.0% (71/323) in those who received UFH during the initial treatment period for their VTE (odds ratio, 0.61; 95% CI, 0.40–0.93). In contrast, among patients with VTE who did not have cancer (odds ratio, 0.94; 95% CI, 0.60–1.47), the 3-month mortality rate was 2.6% (39/1481) for those in the LMWH group versus 2.8% (41/1471) for those in the UFH group.

Adjustment for the primary site of the cancer, age, and gender was performed among data from 3 trials involving 1921 patients, of whom 405 had cancer. In patients with VTE who had cancer, the odds ratio was 0.62 (95% CI 0.36–1.08) for those treated with LMWH, whereas no differences in thrombotic or bleeding complications were noted between those treated with LMWH and those treated with UFH.

Further adjustment for histology, tumour stage, sites of metastases, duration of cancer, and cancer treatment could be made in one trial with 1021 patients, in which 198 patients with cancer had been included. Of the 103 patients with cancer who were treated with LMWH, 19 died within 3 months after study entry, compared with 25 of 95 patients who had cancer in the UFH arm. Without correction for these prognostic factors, the odds ratio was 0.63 (95% CI, 0.32–1.24). After adjusting for histology (adenocarcinoma), stage of the disease (TNM), metastases and site of the tumour, duration of the disease (<1 year; 1–2 years), and treatment, the odds ratio became 0.39 (95% CI, 0.15–1.02), indicating that these adjustments did not change the originally observed beneficial effects of LMWH.

Of course these analyses have several limitations. First, they concern studies with another objective, which was treatment of acute venous thrombotic disease, and the follow-up period was usually limited to 3 months. Second, only patients with cancer who had symptomatic VTE were included, and therefore the relevance for those without VTE remained uncertain. Furthermore, two heparins were compared and therefore the benefit of LMWH alone is unclear. Finally, the types of malignancies varied widely. Nevertheless, the effects were so impressive that randomized controlled trials of the effect of LMWH on cancer progression in patients without VTE were subsequently initiated.

Heparin in patients with cancer without VTE

Six clinical trials have been published studying heparin in patients with cancer without VTE. The inhibitory effects of heparin on cancer growth were studied for the first time in a randomized controlled trial published in 1994, in which 277 patients with both limited and extensive small cell lung cancer were included. Patients were randomized to receive either a prophylactic dose of UFH for 5 weeks or no heparin in addition to their chemotherapy. The response rate was 37% in patients receiving UFH compared with 23% in those receiving only chemotherapy ( P = .004). The median survival time was 317 days in the UFH group compared with 261 days in the no-UFH group ( P = .01). During long-term follow-up (ie, 1, 2, and 3 years after study entry), the survival rates were better in patients receiving UFH, which was statistically significant for those with limited small cell lung cancer.

Altinbas and colleagues also randomized 84 patients with both limited and extensive small cell lung cancer to either a prophylactic dose of a LMWH (dalteparin) or placebo in combination with chemotherapy for a maximum of 18 weeks. The response rate was 69.2% in the patients treated with LMWH, with an overall median survival of 13.0 months, versus 42.5% and 8.0 months, respectively, in the no-LMWH group ( P = .07 and P = .01, respectively).

The FAMOUS trial included a heterogeneous group of 385 patients, all of whom had advanced disease (stage III or IV) and a minimum life expectancy of 3 months. Patients were randomized to receive either a prophylactic dose of a LMWH (dalteparin) or placebo for 12 months with no restriction on concomitant chemotherapy or radiotherapy. At 1, 2, and 3 years, a trend for a better survival was observed in the LMWH group; however, this was significant in a post-hoc analysis in patients who were alive at 17 months after study entry.

A subgroup of patients with cancer who had better survival at study entry (>6 months) was predefined in the MALT study published in 2006. In this randomized trial, 302 patients with different types of advanced cancer that could not be treated curatively and who had a minimum life expectancy of 1 month were randomized to receive either a LMWH (nadroparin; 2 weeks therapeutic dose followed by 4 weeks of a prophylactic dose) or placebo for 6 weeks. The overall median survival was 8.0 months in the nadroparin recipients versus 6.6 months in the placebo group (overall hazard ratio [HR], 0.75; 95% CI, 0.59–0.96). However, in the subgroup of patients with limited disease, defined as an expected survival of at least 6 months at the start, the median survival was 15.4 months versus 9.4 months in the LMWH group and the placebo group, respectively (HR, 0.64; 95% CI, 0.45–0.90).

Sideras and colleagues recruited 141 patients with different types of advanced cancer, a minimum life expectancy of 12 weeks, and an Eastern Cooperative Oncology Group performance status of 0 to 2 to either a prophylactic dose of a LMWH (dalteparin) or placebo or no intervention. The duration of heparin treatment is unclear, but study outcomes included mortality at 1, 2, and 3 years. However, no clinically or statistically significant effect was shown.

In the most recent published study among 69 patients with advanced pancreatic cancer, Icli and colleagues combined LMWH with standard anticancer treatment in 35 patients. Although a clear survival benefit was observed for those treated with LMWH, this study was not randomized and therefore these results should be interpreted with caution.

Heparin in patients with cancer without VTE

Six clinical trials have been published studying heparin in patients with cancer without VTE. The inhibitory effects of heparin on cancer growth were studied for the first time in a randomized controlled trial published in 1994, in which 277 patients with both limited and extensive small cell lung cancer were included. Patients were randomized to receive either a prophylactic dose of UFH for 5 weeks or no heparin in addition to their chemotherapy. The response rate was 37% in patients receiving UFH compared with 23% in those receiving only chemotherapy ( P = .004). The median survival time was 317 days in the UFH group compared with 261 days in the no-UFH group ( P = .01). During long-term follow-up (ie, 1, 2, and 3 years after study entry), the survival rates were better in patients receiving UFH, which was statistically significant for those with limited small cell lung cancer.

Altinbas and colleagues also randomized 84 patients with both limited and extensive small cell lung cancer to either a prophylactic dose of a LMWH (dalteparin) or placebo in combination with chemotherapy for a maximum of 18 weeks. The response rate was 69.2% in the patients treated with LMWH, with an overall median survival of 13.0 months, versus 42.5% and 8.0 months, respectively, in the no-LMWH group ( P = .07 and P = .01, respectively).

The FAMOUS trial included a heterogeneous group of 385 patients, all of whom had advanced disease (stage III or IV) and a minimum life expectancy of 3 months. Patients were randomized to receive either a prophylactic dose of a LMWH (dalteparin) or placebo for 12 months with no restriction on concomitant chemotherapy or radiotherapy. At 1, 2, and 3 years, a trend for a better survival was observed in the LMWH group; however, this was significant in a post-hoc analysis in patients who were alive at 17 months after study entry.

A subgroup of patients with cancer who had better survival at study entry (>6 months) was predefined in the MALT study published in 2006. In this randomized trial, 302 patients with different types of advanced cancer that could not be treated curatively and who had a minimum life expectancy of 1 month were randomized to receive either a LMWH (nadroparin; 2 weeks therapeutic dose followed by 4 weeks of a prophylactic dose) or placebo for 6 weeks. The overall median survival was 8.0 months in the nadroparin recipients versus 6.6 months in the placebo group (overall hazard ratio [HR], 0.75; 95% CI, 0.59–0.96). However, in the subgroup of patients with limited disease, defined as an expected survival of at least 6 months at the start, the median survival was 15.4 months versus 9.4 months in the LMWH group and the placebo group, respectively (HR, 0.64; 95% CI, 0.45–0.90).

Sideras and colleagues recruited 141 patients with different types of advanced cancer, a minimum life expectancy of 12 weeks, and an Eastern Cooperative Oncology Group performance status of 0 to 2 to either a prophylactic dose of a LMWH (dalteparin) or placebo or no intervention. The duration of heparin treatment is unclear, but study outcomes included mortality at 1, 2, and 3 years. However, no clinically or statistically significant effect was shown.

In the most recent published study among 69 patients with advanced pancreatic cancer, Icli and colleagues combined LMWH with standard anticancer treatment in 35 patients. Although a clear survival benefit was observed for those treated with LMWH, this study was not randomized and therefore these results should be interpreted with caution.

What is the relationship between cancer and the activation of hemostasis?

Cancer cells can express several coagulation proteins, thereby activating the coagulation system at several levels. Tissue factor, the initiator of the coagulation system, is expressed by several cancer types. Furthermore, chemokines, such as tumour necrosis factor α, are excreted by cancer cells and can activate monocytes and macrophages, resulting in tissue factor expression on these cells. Lower levels of activated protein C, a natural anticoagulant protein, are observed in patients with cancer because of down-regulation of thrombomodulin on the endothelial cells by these released chemokines, thereby further contributing to the hypercoagulable state. In addition, cancer cells can express urokinase-type and tissue-type plasminogen activators, plasminogen-activator inhibitor 1 and 2, and the plasminogen-activator receptor. The balance among these fibrinolytic proteins leads to an impaired fibrinolysis.

A more recent factor contributing to the hypercoagulable state in patients with cancer is circulating microparticles. Different types of cells, including cancer cells and platelets, are able to form these cell membrane–derived particles. Their phospholipid membrane can bind several coagulation factors. Particles derived from activated monocytes and endothelial and cancer cells can express tissue factor, making them procoagulant. In patients with cancer, high levels of these microparticles have been associated with the occurrence of venous thrombosis and a worse prognosis.

The formation of the end product of the coagulation system, fibrin, around the cancer cells supports angiogenesis and protects the cancer cells against immune attacks.

Furthermore, individual coagulation proteins, irrespective of their role in fibrin formation, create a favorable environment for cancer growth. Binding of FVIIa to the extracellular domain of tissue factor increases the amount of intracellular calcium. This increase activates protein kinase C, which in turn phosphorylates the cytoplasmic tail of tissue factor, which subsequently enhances cancer cell motility and migration and the production of growth factors.

Thrombin stimulates the expression of adhesion molecules on cancer cells and the production of chemokines through binding to protein activated receptors (PARs), members of the G-coupled receptors. Four PAR receptors have been described. PAR signalling increases cancer cell motility and survival and the production of growth factors. Thrombin mainly binds to PAR-1 and PAR-4, whereas other coagulation factors, such as the TF-FVIIa complex and FXa, can activate PAR-2. Furthermore, thrombin activates platelets by binding to their PAR-1 receptor. The activated platelets express P-selectin, a ligand for the adhesion of cancer cells. A shield of platelets is formed and protects cancer cells in the bloodstream from the immune system.

The expression of adhesion molecules on endothelial cells is upregulated by thrombin. Adhesion of cancer cells to the activated platelets and the endothelium facilitates extravasation. Finally, large amounts of growth factors are released by activated platelets, especially vascular endothelial growth factor. In addition to its well-known angiogenic stimulating properties, this hormone causes leakage of plasma proteins through the vessel wall, including fibrinogen, creating again a proangiogenic environment.