Anticoagulants in Thrombosis and Cancer: The Missing Link

Shaker A. Mousa, Ph.D., MBA, FACC, FACB1

ABSTRACT

Many cancer patients reportedly have a hypercoagulable state, with recurrent thrombosis due to the impact of cancer cells and chemotherapy on the coagulation cascade. Studies have demonstrated that unfractionated (UFH) or low-molecular-weight heparin (LMWH) interferes with various processes involved in tumor growth and metasta- sis.These processes might include fibrin formation, binding of heparin to angiogenic growth factors such as basic fibroblast growth factor and vascular endothelial growth factor, modula- tion of tissue factor, and other mechanisms. Clinical trials have indicated a clinically relevant effect of LMWH as compared with UFH on the survival of cancer patients with . Similarly, the impact of on the survival of cancer patients with throm- boembolic disorders was demonstrated. Recent studies from our laboratory defined the role of an LMWH (tinzaparin), warfarin, anti–factor VIIa, and recombinant tissue factor path- way inhibitor in the modulation of angiogenesis, tumor growth, and tumor metastasis.

KEYWORDS: Heparin, LMWH, warfarin, TF/VIIa, TFPI, FGF2, VEGF, cancer, coagulation, platelet, angiogenesis, venous thrombosis

Objectives: Upon completion of this article, the reader should be able to (1) list therapeutic agents that might be effective in reduc- ing the thrombotic risk of cancer patients, and (2) explain some of the effects that these compounds may have on tumor evolution. Accreditation: Tufts University School of Medicine is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. TUSM takes full responsibility for the content, quality, and scientific integrity of this continuing education activity. Credit: Tufts University School of Medicine designates this education activity for a maximum of 1.0 hours credit toward the AMA Physicians Recognition Award in category one. Each physician should claim only those hours that he/she actually spent in the edu- cational activity.

HEPARIN AND phylaxis and treatment of arterial and venous throm- LOW-MOLECULAR-WEIGHT HEPARIN botic disorders, surgical anticoagulation, and interven- Despite the research and development efforts in newer tional use. It is the understanding of the structure of , unfractionated heparin (UFH) and low- heparin that led to the development of LMWHs, syn- molecular-weight heparin (LMWH) will continue to thetic heparinomimetics, , and anti–factor play a pivotal role in the management of thrombotic Xa (anti-Xa) agents.1–4 disorders. Although bleeding and heparin-induced As heparin was discovered over a half-century thrombocytopenia are major side effects of this drug, it ago, our knowledge of the chemical structure and mole- has remained the of choice for the pro- cular interactions of this fascinating polycomponent

Regulation of Tumor Growth by Coagulation-Reactive Drugs; Editor in Chief, Eberhard F. Mammen, M.D.; Guest Editor, Leo Zacharski, M.D. Seminars in Thrombosis and Hemostasis, volume 28, number 1, 2002. Address for correspondence and reprint requests: Shaker A. Mousa, Ph.D., FACC, Albany College of Pharmacy, Albany, NY. E-mail: [email protected]. 1Albany College of Pharmacy, Albany, New York. Copyright © 2002 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662. 0094–6176, p;2002,28,01,45,52,ftx,en;sth00769x. 45 46 SEMINARS IN THROMBOSIS AND HEMOSTASIS/VOLUME 28, NUMBER 1 2002

was limited at the early stages of its development. tumor before signs and symptoms of the tumor itself Through the efforts of major multidisciplinary groups become obvious.20 of researchers and clinicians, it is now well recognized Hemostatic abnormalities are present in a majority that heparin has multiple sites of action and can be used of patients with metastatic cancer. These abnormalities in multiple indications. We may witness the impact of can be categorized as (1) increased platelet aggregation these drugs on the management of various diseases in and activation, (2) abnormal activation of the coagulation the not too distant future. cascade, (3) release of inhibitor 1 is an LMWH produced by (PAI-1), and (4) decreased hepatic synthesis of anticoag- controlled enzymatic depolymerization of conventional, ulant proteins such as and AT. The abnormal unfractionated porcine heparin.3,5,6 In clinical trials, tin- activation of the coagulation cascade is mediated through zaparin was more effective than UFH as treatment for release of tissue factor and other procoagulants from the deep vein thrombosis (DVT),7 was effective in the plasma membrane vesicles of tumor cells22,25 (Fig. 1). treatment of pulmonary embolism8 and the prevention Increasing evidence suggests that thrombotic ep- of DVT in abdominal surgery patients,9 and was supe- isodes may also precede the diagnosis of cancer by rior to warfarin as thromboembolism prophylaxis in months or years, thus representing a potential marker subjects undergoing orthopedic joint (hip or knee) re- for occult malignancy.20 Emphasis has been given to the placement surgery.10 It is also an effective anticoagulant potential risk of cancer therapy (both surgery and for hemodialysis extracorporeal circuits.11 chemotherapy) in enhancing the risk for thromboem- Anti-Xa activity has served as the primary bio- bolic disease.22,25 Postoperative DVT is indeed more marker for assessing the exposure of tinzaparin and other frequent in patients operated for malignant diseases LMWHs. It is used to define the in vitro potency and to than for other disorders. On the other hand, both monitor therapeutic response.12 Given that LMWHs are chemotherapy and hormone therapy are associated with polycomponent moieties with multiple biological actions an increased thrombotic risk, which can be prevented by and distinct time courses, the true pharmacokinetic be- low-dose oral anticoagulation.26,27 In particular, proco- havior of these agents cannot be assessed with assays de- agulant activities of tumor cells have been extensively veloped for a single pharmacological activity. The ab- studied; one of these specific cancer procoagulants solute bioavailability is approximately 90% based on could represent a novel marker of malignancy. anti-Xa activity13 and 93% based on plasma tissue factor pathway inhibitor (TFPI) (data not shown). Clinical trials in which LMWHs with distinct in TREATMENT OF VENOUS vitro potency (anti-Xa/anti-IIa ratio) and ex vivo anti- THROMBOEMBOLISM Xa and anti-IIa activities were tested in patients with IN CANCER PATIENTS DVT following hip replacement found no difference in The management of DVT and efficacy or safety measures as compared with UFH14,15 (PE) in patients with cancer can be a clinical dilemma. despite distinct differences in biomarker activity pro- Comorbid conditions, warfarin failure, difficult venous files. However, anti-Xa activity is sensitive as an indica- access, and a high bleeding risk are some of the factors tor of molecular weight distribution differences with that often complicate anticoagulant therapy in these pa- various heparin fractions.16 LMWHs vary in their af- tients. In addition, the use of central venous access de- finity for antithrombin (AT), presumably as a result of vices is increasing but the optimal treatment of cathe- production method.17 Such differences have been cited ter-related thrombosis remains controversial. UFH is as explaining, in part, the differences in LMWH phar- the traditional standard for the initial treatment of ve- macodynamics as assessed by anti-Xa activity and one nous thromboembolism (VTE), but LMWHs have reason why they cannot be used interchangeably. In been shown to be equally safe and effective in hemody- contrast, TFPI, a vascular endothelial biomarker, might namically stable patients. For long-term treatment or represent a greater potential for the role of LMWH in secondary prophylaxis, vitamin K antagonists remain various diseases.18,19 the mainstay of treatment. However, the inconvenience and narrow therapeutic window of oral anticoagulants make extended therapy unattractive and problematic. THROMBOSIS AND CANCER As a result, LMWHs are being evaluated as an alterna- The etiology of thrombosis in malignancy is multifacto- tive for long-term therapy.26,27 The role of inferior vena rial, and mechanisms include release of procoagulants cava filters in cancer patients remains ill defined, but by tumor cells plus other predisposing hypercoagulable these devices remain the treatment of choice in patients states mediated by chemotherapeutic and radiothera- with contraindications for anticoagulant therapy. peutic agents.20–25 Unexplained thromboembolism may A growing body of evidence has provided a con- be an early indicator of the presence of a malignant vincing demonstration of a strong association between ANTICOAGULANTS IN THROMBOSIS AND CANCER/MOUSA 47

Figure 1 Tumor-mediated hypercoagulable and prothrombotic states. cancer and VTE (Table 1). Patients with cancer are at a trates the positive feedback loop between tumor and remarkably higher risk of VTE than patients free from clot in magnifying each other. malignant disorders during prolonged immobilization Tumor fibrin is a consistent feature of tumor for any cause and following surgical interventions. stroma and is deposited shortly after tumor cell inocula- Standard heparin in adjusted doses or a LMWH in tion.32 As there are several ways in which fibrin may be doses commonly recommended for high-risk surgical beneficial to tumor growth, it is possible that the ability patients represents the prophylactic treatment of choice of normal or malignant tissue to generate fibrin may in- for cancer patients undergoing an extensive abdominal fluence metastasis.32 Many normal tissues and tumor or pelvic intervention.26,27 In cancer patients affected by cells possess a procoagulant activity that is due to a DVT, the treatment with LMWH has been reported to complex of tissue factor and factor VII.22,25,33 reduce mortality to a greater extent than the standard heparin therapy. Such an observation suggests that these agents might modify tumor growth progression ANGIOGENESIS directly or indirectly.28,29 Angiogenesis is a process that is dependent upon coordi- Studies have provided convincing evidence for nate production of angiogenesis stimulatory and in- increased incidences of newly diagnosed malignancy hibitory (angiostatic) molecules, and any imbalance in among patients with unexplained VTE during the first this regulatory circuit might lead to the development of a 6–12 months after the thromboembolic event (see Table number of angiogenesis-mediated diseases (Table 2). 1 for tumors associated with VTE ).28–31 Figure 2 illus- Angiogenesis is a multistep process including activation, adhesion, migration, proliferation, and transmigration of endothelial cells across cell matrices to or from new capil- Table 1 Tumor Type Associated laries and from existing vessels. Angiogenesis is a process with Venous Thrombolism involved in the formation of new vessels by sprouting from preexisting vessels. In contrast, vessel rudiments Pancreatic tumors may organize in place, a process termed vasculogenesis. Mucin-secreting adenocarcinoma from the gastrointestinal Endothelial heterogeneity and organ specificity might system contribute to differences in the response to different an- Lung carcinoma tiangiogenic mechanisms (cultured endothelial cells vs. Ovarian carcinoma microvascular endothelial cells isolated from different tis- Endometrial carcinoma sues). Under normal physiological conditions, in a mature Intracranial tumors organism endothelial cell turnover is extremely slow Acute promyelocytic leukemia (months to years); this is in contrast to the fast turnover Myeloproliferative disorders of endothelial cells under pathological angiogenesis 48 SEMINARS IN THROMBOSIS AND HEMOSTASIS/VOLUME 28, NUMBER 1 2002

Figure 2 Interplay between tumor and clot illustrating a positive feedback loop. Tumor promotes hypercoagulable state and clot (platelet-fibrin clot) induces tumor growth and tumor angiogenesis.

conditions. However, angiogenesis can be activated for a limited time in certain situations such as wound healing and ovulation. Table 2 Diseases Associated with Angiogenesis In certain pathological states, such as human me- Prevalence tastasis33 and ocular neovascularization disorders includ- Disease (or Incidence*) ing diabetic retinopathy and age-related macular degen- Solid tumor cancer >600,000 eration, there is excessive and sustained angiogenesis. Lung, breast, prostate, colon, renal, Hence, understanding the mechanisms involved in the bladder, pancreatic, glioblastomas, regulation of angiogenesis could have a major impact in neuroblastomas, and others the prevention and treatment of pathological angiogene- Ocular ailments sis processes. In addition, endothelial cells play a major Macular degeneration 650,000 role in the modeling of blood vessels. The interplay of Diabetic retinopathy 300,000 growth factors, cell adhesion molecules, and specific sig- Corneal transplant 100,000 nal transduction pathways either in the maintenance of Myopic degeneration 200,000 the quiescent state or in the reactivation of endothelium Inflammatory diseases is critical in physiological and pathological angiogenesis Arthritis 2.1 million processes. A combined defect in the overproduction of Psoriasis 3.0 million positive regulators of angiogenesis and a deficiency in en- IBD/other chronic IDs† >2.0 million dogenous angiostatic mediators are a feature documented *New cases per year. in tumor angiogenesis, psoriasis, rheumatoid arthritis, †IBD, inflammatory bowel disease; ID, inflammatory diseases. and other neovascularization-mediated disorders. Hence, ANTICOAGULANTS IN THROMBOSIS AND CANCER/MOUSA 49 understanding the mechanisms involved in the regulation endothelial tube formation and in vivo angiogenesis of angiogenesis could have a major impact in the treat- mediated by angiogenic factors and cancer cells, were ment of pathological angiogenesis.34 demonstrated. The in vivo effects of those different an- ticoagulants on angiogenesis in the chick chorioallan- toic membrane (CAM) model were determined. Angiogenesis-Modulating Twenty-four hours after stimulating angiogenesis on Mechanisms and Agents the CAM with basic fibroblast growth factor (FGF2), The origin of the mechanisms and agents involved in lipopolysaccharide (LPS), or colon carcinoma (HCT- angiogenesis modulation35–40 includes the following: 116), tinzaparin, warfarin, anti-VIIa, or r-TFPI were di- rectly applied to the growth factor–saturated filter disk 1. Extracellular matrix (ECM): collagen (endostatin), or injected intravenously into the embryonic circula- fibronectin, or vitronectin (RGD analogues). tion. Data demonstrated significant and comparable in- 2. Cell adhesion receptors: v3, 51, and 11. hibitory effects of the LMWH tinzaparin, anti-VIIa, 3. Fibrinolytic system: kininogen domain 5 and plas- and r-TFPI in a concentration-dependent manner on minogen-driven angiostatin. endothelial cell tube formation. Tinzaparin, warfarin, 4. Coagulation system: heparin-like glycosaminogly- anti-VIIa, and r-TFPI blocked FGF2-induced angio- cans (GAGS) and fibrin. genesis in the CAM model by 80–100%. In addition, 5. Platelet derived: platelet-derived growth factor significant inhibition of colon or lung carcinoma–induced (PDGF), platelet factor 4 (PF4), thrombospondin, angiogenesis (Fig. 3), tumor growth (Fig. 4), and re- and thrombomodulin. gression was demonstrated with tinzaparin, warfarin, 6. Matrix-degrading enzymes: MMPs. anti-VIIa, and r-TFPI. These studies demonstrated a significant role for tinzaparin, warfarin, anti-VIIa, and tinzaparin-releasable TFPI in the regulation of angio- ACTIVATION OF COAGULATION genesis and tumor growth.37,43 Thus, modulation of tis- AND ANGIOGENESIS IN CANCER sue factor/VIIa noncoagulant activities by those differ- Tissue factor has been implicated in the upregulation of ent agents might be a useful therapeutic approach for proangiogenic factors such as vascular endothelial the inhibition of angiogenesis associated with human growth factor (VEGF) by tumor cells. This is due to a tumor growth and inflammatory diseases. See Table 2 complex interaction between tumor cells, macrophages, for diseases associated with pathological angiogenesis. and endothelial cells leading to tissue factor expression, fibrin formation, and tumor angiogenesis.41,42 LMWH,TFPI, and Tumor Dissemination The significance of tissue factor in cancer biology is sug- ANTICOAGULANTS IN gested by studies reporting its involvement in metastasis THE MODULATION OF ANGIOGENESIS and angiogenesis. Tinzaparin is an LMWH that is pro- The effects of the LMWH tinzaparin, warfarin, anti- duced by heparinase depolymerization of UFH allowing VIIa, and recombinant TFPI (r-TFPI) on the modula- for its relatively high sulfate-to-carboxylate ratio. Beyond tion of angiogenesis-related processes, including in vitro its potent plasmatic effects on AT-inhibitable coagula-

Figure 3 Antiangiogenesis efficacy of tinzaparin on tumor-induced angiogenesis. 50 SEMINARS IN THROMBOSIS AND HEMOSTASIS/VOLUME 28, NUMBER 1 2002

Figure 4 Effect of a single tinzaparin (0.1 mg) treatment on colon (left) and fibrosarcoma (right) growth in the CAM model after 1 week.

tion factors, tinzaparin is a very effective LMWH in Hemostasis and Tumor Dissemination causing the release of TFPI from the endothelial cells, Platelets and fibrin are key catalysts for tumor adhesion, the natural inhibitor of tissue factor procoagulant and survival, and metastasis. Tumor cell metastasis and noncoagulant effects. The present study was undertaken thrombosis are the major causes of death in cancer pa- to investigate the effect of the LMWH tinzaparin as well tients.31,46,47 as recombinant TFPI on experimental lung metastasis. Using the B16 melanoma injectable model of metastasis, we found that subcutaneous injection of tin- SELECTED ANGIOGENESIS INHIBITOR zaparin (10 mg/kg) 4 hours before intravenous injection STRATEGIES UNDER INVESTIGATION of 2.5 105 melanoma cells reduced lung tumor for- Recent evidence suggests that, in spite of the redun- mation in experimental mice by 89% (31 ± 23 vs. 3 ± 2, dancy of angiogenic factors potentially involved in P < 0.001). In a second experimental group, in addition pathological angiogenesis, strategies aimed at antago- to the initial (pre-tumor cell) dose, subcutaneous tinza- nizing one specific endothelial cell mitogen at its release parin (10 mg/kg) was administered daily for 14 days, at or receptor levels may form the basis for an effective and which time lung seeding was assessed. In the latter safe treatment of various angiogenesis-mediated disease group, lung tumor formation was reduced by 96% (P < processes (Table 3).48,49 0.001). No bleeding problems were observed in any of Tumors are dynamic, complex, living tissues un- the heparinized animals. In order to determine the anti- dergoing the varied processes of tissue growth under the coagulant activity of tinzaparin, 4 hours after a single guidance of aberrant malignant cells. Cytotoxic anti- subcutaneous dose, whole blood recalcification was measured using a Sonoclot® analyzer. Tinzaparin (10 mg/kg subcutaneously) prolonged the clotting time Table 3 Examples of Different Antiangiogenesis fourfold. Furthermore, measuring the platelet count (a Strategies under Investigation sensitive marker of intravascular coagulation) before VEGF receptor–tyrosine kinase antagonists, Tie1, Tie2 and 15 minutes after determined the effect of tinzaparin (SU-5416) on tumor cell–induced clotting activation in vivo after Anti-VEGF antibody tumor cell injection in control and tinzaparin-treated Humanized form of LM609 (Vitaxin) animals. Following intravenous injection of 2 106 Small molecule v3 antagonists tumor cells, a rapid and significant fall in platelet count TNP470 was observed (from 939 37 106/mL to 498 2-Methoxyestradiol 94 106/mL, P < 0.01). In tinzaparin-treated animals, CM-101 a significant reversal to normal platelet count was Shark cartilage extract (AE-941) achieved (921 104 106/mL). Intravenous injection Thalidomide and thalidomide analogues of TFPI (700 ng) 5 minutes prior to tumor cell injection Endostatin and angiostatin also reduced B16 lung metastasis (85%, P < 0.01) and High-molecular-weight kininogen domain 5 abolished tumor cell–induced thrombocytopenia. Our Matrix metalloprotinease inhibitors (AG3340) results support the potential role of the LMWH tinza- receptor antagonist parin and its releasable TFPI in tumor metastasis.44 ANTICOAGULANTS IN THROMBOSIS AND CANCER/MOUSA 51 cancer therapies have focused solely on the eradication 1. Subcutaneous LMWH plus low-dose oral warfarin of the malignant cell, which is an absolute necessity; might provide improved efficacy and safety. The ad- however, even the most heroic therapeutic strategies dition of a daily might provide additional rarely achieve cure of many tumor types. The recogni- benefits in some tumor types. tion that the growth processes of tumors are normal 2. Long-term (6 months) versus short-term (6 weeks) processes, that the invasion processes of tumors are nor- use of the preceding regimen. mal processes, and that it is inappropriate activation of 3. Efficacy of LMWH versus warfarin on different tumor the processes that results in the morbidity of malignant types and at different stages of tumor progression. disease allows the elucidation of a broad spectrum of 4. Interactions of anticoagulant with chemotherapeutic new therapeutic targets in cancer. The integration of agents and with other antiangiogenesis strategies. antiangiogenic agents into existing cancer chemother- apy regimens might lead to improved efficacy and safety for many standard cytotoxic therapies. REFERENCES

1. Weitz JI. Low-molecular-weight heparin. N Engl J Med Adjuvant use of 5-Fluorouracil and 1997;337:688–698 LMWH in Proliferative Vitreoretinopathy 2. Aguilar D, Goldhaber SZ. Clinical uses of low molecular A recent investigation assessed the safety and efficacy weight heparin. Chest 1999;115:1418–1423 of adjuvant combination therapy using 5-fluorouracil 3. Linhardt RJ, Gunay NS. Production and chemical processing of low molecular weight . Semin Thromb Hemost (5-FU) and LMWH for prevention of proliferative vit- 1999;25:5–16 50 reoretinopathy (PVR) after vitrectomy and retinal 4. Mousa SA, Fareed JW. Advances in anticoagulant, an- reattachment surgery. The study was a prospective ran- tithrombotic and thrombolytic drugs. Exp Opin Invest Drugs domized, double-masked, placebo-controlled trial. The 2001;10:157–162 participants were 174 high-risk patients randomized to 5. Nader HB, Walenga JM, Berkowitz SD, et al. Preclinical dif- receive either 5-FU and LMWH therapy or placebo. ferentiation of low molecular weight heparins. Semin Thromb Patients were selected from all patients undergoing pri- Hemost 1999;25:63–72 6. Mousa SA. Comparative efficacy among different low molec- mary vitrectomy for rhegmatogenous retinal detach- ular weight heparins (LMWHs) and drug interaction: impli- ment. Standard surgery with 5-FU and LMWH ther- cations in the management of vascular disorders. Thromb apy or placebo was compared at the 6-month follow-up. Haemost 2000;26(Suppl 1):39–46 The main outcome measures were development of post- 7. Hull RD, Raskob GE, Pineo GF, et al. Subcutaneous low- operative PVR, retinal reattachment at 6 months after molecular weight heparin compared with continuous intra- surgery, single operation reattachment rate, number of venous heparin in the treatment of proximal-vein thrombosis. reoperations, and best corrected visual acuity. There N Engl J Med 1992;326:975–982 8. Simonneau G, Sors H, Charbonnier B, et al. A comparison of were 87 patients in the 5-FU and LMWH therapy low-molecular-weight heparin with unfractionated heparin group and 87 in the placebo group. The incidence of for acute pulmonary embolism. N Engl J Med 1997;337: postoperative PVR was significantly lower (P = 0.02) in 663–669 the 5-FU and LMWH therapy group compared with 9. Hull RD, Raskob GE, Pineo GF, et al. A comparison of sub- the placebo group. There was a significant reduction cutaneous low-molecular weight heparin with warfarin so- in the incidence of postoperative PVR in patients re- dium for prophylaxis against deep-vein thrombosis after hip ceiving the 5-FU and LMWH therapy and in the reop- or knee implantation. N Engl J Med 1993;329:1370–1376 10. Leizorovicz A, Picolet H, Peyrieux JC, Boissel JP, and the eration rate resulting from PVR. This trial shows that HBPM research group. Prevention of perioperative deep vein the incidence of PVR can be reduced with 5-FU and thrombosis in general surgery: a multicenter double blind LMWH.50 study comparing two doses of logiparin and standard heparin. Br J Surg 1991;78:412–416 11. Ryan KE, Lane DA, Flynn A, et al. Dose finding study of a ANTIANGIOGENESIS AGENTS IN low molecular weight heparin, Innohep, in haemodialysis. Thromb Haemost 1991;66:277–282 MALIGNANCY AND HEMOSTASIS 12. Troy S, Fruncillo R, Ozawa T,et al. Absolute and comparative Recent evidence suggests enhanced thrombogenicity subcutaneous bioavailability of , a low mole- and fibrinolytic deficit with different angiogenesis in- cular weight heparin. Thromb Haemost 1997;78:871–875 hibitor agents.50,51 The exact mechanism for the inter- 13. Pedersen PC, Østergaard PB, Hedner U, Bergqvist D, face between angiogenesis modulation mechanisms and Mätzsch T. Pharmacokinetics of low molecular weight hepa- hemostasis is unclear at this point. rin, logiparin, after intravenous and subcutaneous administra- tion to healthy volunteers. Thromb Res 1991;61:477–487 The following preclinical and clinical investiga- 14. Mätzsch T,Bergqvist D, Hedner U, Østergaard PB. Effects of tions should provide information needed to optimize an enzymatically depolymerized heparin as compared with the benefit of anticoagulants in oncology and to better conventional heparin in healthy volunteers. Thromb Haemost understand the mechanisms of the anticoagulant effect: 1987;57:97–101 52 SEMINARS IN THROMBOSIS AND HEMOSTASIS/VOLUME 28, NUMBER 1 2002

15. Bara L, Planes A, Samama M-M. Occurrence of thrombosis 33. Falanga A. Mechanisms of hypercoagulation in malignancy and haemorrhage, relationship with anti-Xa, anti-IIa activi- and during chemotherapy. Haemostasis 1998;28:50–55 ties, and D-dimer plasma levels in patients receiving low mol- 34. Mousa SA. Mechanisms of angiogenesis in vascular disor- ecular weight heparin, enoxaparin or tinzaparin, to prevent ders: potential therapeutic targets. In: Mousa S, Landes RG, deep vein thrombosis after hip surgery. Br J Haematol eds. Angiogenesis Inhibitors and Stimulators: Potential Ther- 1999;104:230–240 apeutic Implications. Gorgtown, TX: Landes Bioscience; 16. Emmanuele RM, Fareed J. The effect of molecular weight on 2000:1–12 the bioavailability of heparin. Thromb Res 1987;48:591–596 35. Zhang H-T, Harris AL. Anti-angiogenic therapies in cancer 17. Brieger D, Dawes J. Production method affects the pharma- clinical trials. Exp Opin Invest Drugs 1998;7:1629–1655 cokinetic and ex vivo biological properties of low molecular 36. Mousa SA, Mohamed S, Smallheer J, Jadhav PK, Varner J. weight heparins. Thromb Haemost 1997;77:317–322 Anti-angiogenesis efficacy of small molecule a5b1 integrin 18. Mousa SA, Bozarth J, Larnkjaer A, Johanson K. Vascular ef- antagonists, SJ749. Blood 1999;94(Suppl 1):620a, 2755 Abst fects of heparin molecular weight fractions and LMWH on 37. Colman RW, Jameson BA, Lin Y, Mousa SA. Inhibition of the release of TFPI from human endothelial cells. Blood angiogenesis by kininogen domain 5. Blood 2000;95:543–550 2000;16:59, 3928. 38. Mousa SA, Mohamed S. Anti-angiogenesis efficacy of the 19. Mousa SA, Mohamed S. Anti-angiogenesis efficacy of the low molecular weight heparin (LMWH) tinzaparin and tis- low molecular weight heparin (LMWH), tinzaparin and tis- sue factor pathway inhibitor (TFPI). Blood 1999;94(Suppl sue factor pathway inhibitor (TFPI). Blood 1999;94(Suppl 1):22a, 82-I Abst 1):22a, 82 Abst 39. Browder T, Folkman J, Pirie-Shepherd S. The hemostasis sys- 20. GoldbergRJ, Seneff M, Gore JM. Occult malignant neoplasm tem as a regulator of angiogenesis. J Biol Chem 2000; in patients with deep venous thrombosis. Ann Intern Med 275:1521–1524 1987;147:251–253 40. Brooks PC, Clark RAF, Cheresh DA. Requirement of vascu- 21. Baron JA, Gridley G, Weiderpass E, Nyren O, Linet M. lar integrin v3 for angiogenesis. Science 1994;264:569–571 Venous thromboembolism and cancer. Lancet 1998;351: 41. Bell WR. The fibrinolytic system in neoplasia. Semin 1077–1080 Thromb Haemost 1996;22:459–478 22. Rickles FR, Edwards RL. Activation of blood coagulation in 42. Ruf W, Mueller BM. Tissue factor in cancer angiogenesis and cancer: Trousseau’s syndrome revisited. Blood 1982;62:14–31 metastasis. Curr Opin Hematol 1996;3:379–384 23. Levine MN. Prevention of thrombotic disorders in cancer pa- 43. Shoji M, Hancock WW, Abe K, et al. Activation of coagula- tients undergoing chemotherapy. Thromb Haemost 1997;78: tion and angiogenesis in cancer: immunohistochemical local- 133–136 ization in situ of clotting proteins and VEGF in human can- 24. Levine MN, Gent M, Hirsh J, et al. The thrombogenic effect cer. Am J Pathol 1998;152:399–411 of anticancer drug therapy in women with stage II breast can- 44. Mousa SA, Mohamed S. Anti-angiogenesis efficacy and anti- cer. N Engl J Med 1998;318:404–407 tumor effects of warfarin: potential mechanisms. FASEB 25. Kakkar AJ, De Ruvo N, Tebbutt S, Williamson RCN. Extrin- 2001;15:A118, 149 sic pathway activation with elevated tissue factor and factor 45. Amirkhosravi A, Francis J, Mousa SA. Anti-metastatic effect VIIa in patients with cancer. Lancet 1995;346:1004–1005 of low molecular weight heparin (LMWH) tinzaparin and 26. Kakkar AJ, Williamson RC. Prevention of venous throm- tissue factor pathway inhibitor (TFPI). Thromb Haemost boembolism in cancer using low molecular weight heparin. 2001;Suppl:P1409 Abst Haemostasis 1997;27(Suppl 1):32–37 46. Maloney JP, Silliman CC, Ambruso DR, et al. In vitro release 27. Trousseau A. Plegmasia alba dolens. In: Baillier JB, ed. Clin- of VEGF during platelet aggregation. Am J Physiol 1998; ique de l’Hotel-Dieu de Paris, 2nd ed. 1865;3:654–712 275:H1054-H1061 28. Zacharski LR, Ornstein DL. Heparin and cancer. Thromb 47. Francis JL. Haemostasis and cancer. Med Lab Sci 1989; Haemost 1998;80:10–23 46:331–346 29. Gillis S, Dann EJ, Eldor A. A low molecular weight heparin 48. Brower V. Tumor angiogenesis—new drugs on the block. Na- in the prophylaxis and treatment of disseminated intravascu- ture Biotechnol 1999;17:963–968 lar coagulation in acute leukemia. Eur J Haematol 1995:54: 49. Paku S. Current concepts of tumor-induced angiogenesis. 59–60 Pathol Oncol Res 1998;4:62–75 30. Haward W. Phlebitis and thrombosis. Lancet 1960;1: 50. Hassan AR, Hing YK, Catey B, et al. Adjuvant 5-fluorouracil 650–655 and heparin prevents proliferative vitreoretinopathy: results 31. Dvorak HF. Thrombosis and cancer. Hum Pathol from a randomized, double-blind, controlled clinical trial. 1987;18:275–284 Ophthalmology 2001;108:1179–1183 32. Dvorak HF, Senger DR, Dvorak AM. Fibrin as a component 51. Osman K, Comenzo R, Rajkumar SV. Deep vein thrombosis of the tumor stroma: origins and biological significance. Can- and thalidomide therapy for multiple myeloma. N Engl J cer Metastasis Rev 1983;2:41–73 Med 2001;344:1951–1952