Cancerand MetastasisReviewsII: 283-290, 1992. © 1992KluwerAcademic Publishers. Printedin the Netherlands.

The role of fibrin in tumor metastasis Vincenzo Costantini 1 and Leo R. Zacharski 2 t Institute of Internal and Vascular Medicine, Universit~ of Perugia, Perugia, Italy and 2 The Department of Medicine, Dartmouth Medical School and the Department of Veterans Affairs Medical Center, White River Junction, Vermont 05009, USA

Key words: fibrinogen, fibrin, neoplasia, tumor metastasis

Summary A volume of data that has accumulated for over a century has suggested that fibrin may facilitate the persistence and progression of malignancy. Techniques that have been developed recently have shown that fibrin is indeed a component of the connective tissue stroma in human malignancy but in only a few tumor types. However, therapeutic intervention studies with drugs that limit thrombin activity or enhance fibrinolysis have shown favorable clinical effects in at least one such tumor type. These favorable findings affirm the concept that cause-and-effect relationships do, in fact, exist between thrombin generation with fibrin formation and tumor progression, and suggest that a rational basis exists for the design of future drug intervention trials that target reactions relevant to specific tumor types. These findings also provide a basis for the design of experiments capable of defining further the role of fibrin in the integrity of these tumor types. Because fibrinogen is found much more commonly than fibrin in the connective tissue of a variety of human malignancies, attention might reassumably be directed to determining the possible contribution of this molecule as well as of fibrin to tumor progression.

Introduction Fibrin has been linked to cancer biology for well over a century [reviewed in reference 1]. The focus of interest in the fibrin-tumor association has been the possibility that fibrin may form around circulating tumor cells that facilitates microvascular entrapment required for metastasis, and also that fibrin may be a biologically important constituent of the tumor connective tissue stroma. Painstaking studies conducted in the late Nineteenth and early Twentieth Centuries [1] of human tissues obtained at autopsy from cancer patients documented the occurrence of thrombi around embolic tumor cells within blood vessels. Such findings were subsequently documented in tumor-bearing experimental animals [2-5]. In certain of these animal models of malignancy, tumor dissemination could be

blocked by administering drugs that either limited fibrin formation because of their anticoagulant properties or lysed the fibrin because of their ability to convert plasminogen to plasmin [reviewed in references 2-4]. More recently it has been emphasized that fibrin may be a constituent of the connective tissue stroma of both animal [6-10] and human [11, 12] tumors. Such extravascular fibrin is conceived of a possibly enhancing tumor integrity by providing a scaffolding for tumor cell growth, enhancing angiogenesis, or protecting the tumor from attack by host inflammatory cells that would presumably destroy the tumor [6, 7, 10, 13, 14]. The latter has been generalized into the concept that fibrin may serve as a protective 'cocoon' that surrounds the tumor, and that the tumor resembles a 'wound', from the point of view of its fibrin content, but a

284 wound that cannot heal because it paradoxically serves to perpetuate the 'injury', that is, the tumor itself [6]. If true for human malignancy, such a formulation would be fortuitous indeed because such fibrin could presumably be either removed by fibrinolytic enzymes or prevented from being formed by anticoagulants to the advantage of the patient. Unfortunately, however, the true state of affairs seems not to be so simple and straightforward. For example, the composition of the thrombi found long ago to surround intravascular tumor emboli [1] has not been studied in detail and may consist of either platelets [15, 16] or fibrin (or both). The fact that such thrombi have formed does not necessarily mean that entrapment presumed to be necessary for metastasis formation has occurred, but may rather be merely an associated but not causally related phenomenon. To assign primary importance to such thrombi in human metastasis would require studies of effects of intervention with antithrombotic drugs. Such studies have indeed been performed but these have shown beneficial effects in only one tumor type so far [17-22] while the natural history of certain other tumor types is unaffected by this form of therapy [18]. Although such an approach is indeed effective in certain experimental animal models of malignancy [2, 15, 23], these lack resemblance to clinical situations, for example, because treatment of humans has invariably been initiated well after diagnosis whereas treatment usually precedes inoculation of tumor cells in animal models. This means that metastasis has almost certainly already occurred in humans so treated and clinical measures of outcome have correspondingly been improved survival and delayed time to tumor progression rather than inhibition of metastasis per se [17-22]. To complicate matters further, fibrinolytic therapy has been shown to be effective in certain clinical settings [20, 21, 24] while success has been claimed for antifibrinolytic therapy in others [25]. Similar difficulties arise upon closer examination of the literature describing the importance of extravascular fibrin deposition at tumor sites [6-10]. Fibrin undoubtedly is formed in certain experimental animal models but the degree to which hu-

man malignancy corresponds to such models (if at all) has only recently begun to be clarified. There is, therefore, little basis for generalizations concerning the existence of fibrin in human malignancy let alone its possible role in tumor progression. The purposes of this paper are to review recent progress in our understanding of fibrin formation in human tumors in terms of the cellular sites of procoagulants responsible for thrombin generation capable of converting fibrinogen to fibrin, to define criteria for the existence of fibrin as opposed to fibrinogen in situ in tumor tissues, to examine data from clinical trials of anticoagulant fibrinolytic therapy in cancer that are required to establish cause-and-effect relationships between fibrin formation and neoplastic progression, and finally to focus attention on the potential biologic role of either fibrinogen or fibrin in supporting tumor growth to provide a basis for future studies.

Tumor fibrin formation

For tumor fibrin to form, there must be a pathway that results in the formation of thrombin that, in turn, converts fibrinogen to fibrin. For fibrin formation to be verified, criteria must be developed that can be applied to intact tumor tissues as they exist in patients. The assumption that fibrin exists in human tumors cannot be generalized from studies in experimental animals. Furthermore, assumptions based on data obtained from studies on cells in culture or from other in vitro or semipurified systems may not be valid because such conditions may not resemble conditions within complex tissues in vivo. Our approach to this problem was to use immunohistochemical procedures applied to intact human tumor tissues. This approach is useful because high quality antibodies exist to virtually every known clotting factor and fibrinolytic pathway component as well as to both fibrinogen and fibrin. These reagents permitted a systematic search for the existence of each of these elements in tumor tissues and permitted their precise microanatomic localization to either tumor cells, host inflammato-

285 ry cells, or vascular endothelial cells [11]. Fibrinogen was distinguishable from fibrin because of the availability of a monoclonal antibody the reactivity of which required the existence of an intact 14 - 15 bond on the BO chain of fibrinogen. This bond is broken by thrombin proteolysis during conversion of fibrinogen to fibrin. Reactivity of this antibody signalled the presence of fibrinogen. Similarly, a different monoclonal antibody the reactivity of which required cleavage of the same bond was used to identify fibrin. Thus, staining with the latter antibody provided evidence for generation of thrombin that had converted fibrinogen to fibrin [26, 27]. These investigations were rewarded by data indicating a consistent pattern of reactivity among cases within a given tumor type but variability in patterns of reactivity between tumor types. These results have recently been reviewed and are summarized here. Various human tumor types were found to fall into one of three different categories [11, 12]. In the first, the existence of tumor cell clotting factors (procoagulants) was discovered together with conversion of fibrinogen to fibrin in the adjacent connective tissue. Tumors in this category included small cell carcinoma of the lung (SCCL) [28-30], renal cell carcinoma [31, 32], and malignant melanoma [33]. Tumor cell plasminogen activators were either absent (SCCL, renal cell carcinoma), or scant (malignant melanoma). Tumors in the second category lacked an intact tumor cell associated clotting pathway and had no tumor cell associated fibrin formation. Rather, the tumor cells expressed urokinase-type plasminogen activator (U-PA). Tumor types included in this category were non-small cell lung cancer (N-SCLC) [34], breast cancer [35, 36], colon cancer [37], and prostate cancer [38]. A third category of tumors had neither tumor cell procoagulants nor plasminogen activators. These included mesothelioma [39] and lymphomas [40]. The power of this experimental approach is illustrated by the fact that fibrin formation was detected in the tumor vasculature in prostate cancer [38], and adjacent to tumor-associated macrophages in N-SCLC [34] and lymphoma. These macrophages also expressed an intact extrinsic (tissue factor-initiated) pathway of coagulation.

More recently we have used a probe that reacts only with the activated form of factor X (Xa) and discovered Xa on cells in a pattern consistent with the above results [41]. The activity of tumor cell procoagulants in situ is evidently unchecked by tissue factor pathway inhibitor [42]. While factor X and Xa were found consistently in the above pattern [43], certain other clotting factors were not suggesting that the initiator of coagulation might differ between tumor types. The precise identity of these tumor cell coagulation initiators in sites remains to be determined. We concluded that fibrin did, in fact, exist in tumor tissue but only in a restricted number of tumor types [28-33]. By contrast, fibrinogen was observed commonly, even in non-procoagulant tumor types, and was notably abundant in breast cancer [35].

Induction of fibrinolysis and inhibition of fibrin formation as cancer therapy

This subject is reviewed in detail elsewhere in this volume. Of relevance here is the fact that distinctions can now be made between tumor types in terms of their coagulation biology that provide a basis for interpretation of results of completed clinical trials and the design of new clinical trials. For example, it has been shown that SCCL (that expresses tumor cell procoagulants and is associated with fibrin formation [28-33]) responded to therapy with both anticoagulants (warfarin [17-19] and heparin [22]) and fibrinolytic activators [20, 21] while several other tumor types that do not share these properties did not respond to this treatment approach [18, 34, 37, 38]. It might be reasonable to search for other as yet undiscovered procoagulant tumor types, to apply this treatment approach to procoagulant tumor types that have not yet been tested, and to extend studies in SCCL to clinical trials of combinations of anticoagulant plus fibrinolytic drugs or of more effective inhibitors of thrombin generation and activity. Furthermore, while inhibitors of thrombin formation or activity are apparently ineffective i n non-procoagulant tumor types [8, 34, 37, 38], it is theoretically conceivable

286 that fibrinolytic therapy may still be effective because of its presumed ability to digest fibrinogen that is sometimes present in abundance in the tumor connective tissue (for example, in breast cancer [35]). This concept takes into consideration the possibility that fibrinogen as well as fibrin may contribute to tumor integrity.

Basis for implicating fibrin and fibrinogen in tumor biology We have shown above that theoretical constructs that attempt to describe the contribution of fibrin to tumor integrity can now be evaluated in terms of whether or not a known procoagulant pathway capable of generating thrombin and transformation of fibrinogen to fibrin does, in fact, exist in tumor masses. We have found that, for human disease, local tumor cell thrombin generation and fibrin formation are features of a minority of tumor types [28-33]. These findings highlight the potential problems that result from attempting to extrapolate too freely results from experimental animal and in vitro models to human disease. However, they also provide an opportunity to enquire whether any specific individual theory on the role of fibrinogen or fibrin in tumor progression applies to individual tumor types in which either of these really exists. Unfortunately, experimental models of malignancy have almost never been designed based on consideration of defined conditions that are known to exist within a given human tumor type in vivo. For example, a wide variety of transformed cell types can be coaxed into growing beneath the renal capsule in vivo in rodents by encasing them in thrombin-clotted fibrinogen (i.e., fibrin) [44]. However, this model was apparently designed to provide a setting in which chemotherapeutic agents could be tested [45] and little attention has been given to understanding why the fibrin matrix required for the success of this model is so accommodating to tumor cell growth. It would seem that such a model might well be examined from this point of view particularly in terms of specific human tumor types that apparently thrive under similar conditions [28-33].

Further questions come to the fore upon examination of specific data on possible interactions of tumor cells with either fibrinogen or fibrin. For example, cultured endothelial cells orient themselves into tubular, vessel-like channels that surround fibrin strands in vitro [46]. CapiUar growth proceeds into porous chambers that are filled with fibrin following insertion into the subcutaneous tissue of experimental animals [8]. Such evidence has been taken to support a role for fibrin in tumor angiogenesis [6, 7]. However, tumor types that lack fibrin also become vascularized. Thus, angiogenesis must be capable of occurring by other mechanisms and may be fibrin-promoted in only a few tumor types, if at all. What about the concept that either fibrinogen or fibrin provide a matrix for tumor growth? It has been proposed that a fibrin gel may provide a solid 'surface' or interface within the living organism [47] that is capable of guiding cell growth and tissue remodelling. However, the notion that fibrin provides a 'scaffolding' for cancer growth in general finds no direct support from experimental observations. To the contrary, neoplastic transformation has been found to be associated with reduced ability of the cells to interact with fibrin [48--50]. Furthermore, in models of normal cell migration, such motility may be either increased or decreased by either fibrinogen or fibrin depending on experimental conditions [51-53]. While certain tumor cell types in culture have been shown to express surface receptors for fibrinogen [54-56], it is not known whether such receptors exist in vivo. An intriguing possibility remains that a fibrinogen [13] or fibrin [14] coating on tumor cells circulating within blood vessels may protect them from attack by natural killer cells. In experimental settings [14], the effects of drugs that enhance cellular immunity against tumor cells are themselves enhanced by anticoagulants. It has been proposed that this enhancement results from limitation of tumor cell investment with fibrin [14]. It is intriguing that the two human tumor types for which benefit from enhancement of cellular immunity has been claimed, namely renal cell carcinoma [57] and malignant melanoma [58], are also among the few human tumor types that express procoagulants and tumor-associated

287 fibrin [31-33]. Unfortunately, the possibility that administration of anticoagulants might augment the partial benefits realized from immune enhancement therapy has not yet been subjected to clinical testing. Furthermore, it is not known whether the fibrin that can be induced on the surface of tumor cells in experimental settings [14] has a counterpart in humans in vivo. Thus, the composition of the 'thrombi' that are believed to exist adjacent to tumor cells within vessels is open to question. Such thrombi may indeed consist of fibrin in some instances but they may alternatively consist primarily of platelets [15, 16] or other constituents of the blood. More fundamental questions arise concerning the state of the fibrinogen that is a prominent feature of the connective tissue stroma of certain tumor types, notably breast cancer [35]. The impression is that the presence of this fibrinogen may not be explainable simply on the basis of passive diffusion from the plasma. This suspicion is based on the observation that plasma proteins of similar or smaller size are not represented in this location. Thus, the fibrinogen may have been specifically sequestered in place by mechanisms that are not yet understood. For example, this fibrinogen may have been 'gelled' by reactions other than through transformation by thrombin to fibrin. Tissue transglutaminases exist that are capable of stabilizing fibrinogen through gelatin by intermolecular crosslinking [59]. While plasma transglutaminase, that is, fibrin stabilizing factor or factor XIIIa, crosslinks gamma chains on adjacent fibrin molecules, tissue transglutaminase is capable of causing crosslinking between sites on the gamma and alpha chains of fibrinogen [60]. Such stabilized fibrinogen would not be detected as fibrin because existing criteria for fibrin in situ depend on demonstration of exposure of the thrombin cleavage site on fibrinogen [26, 27]. Such functions of tissue transglutaminases are known to contribute to cell anchorage within tissues [59]. The problem is that, in malignancy, it is not so much cell anchorage that is of concern as it is the lack of such anchorage that is reflected in departure of cells from sites of proliferation during invasion and metastasis. Furthermore, it has been shown that transglutaminase ac-

tivity is reduced in proliferating in contrast to resting cells [61] and malignant cells have a high rate of proliferation. Direct examination of tissues for transglutaminase activity has shown that such activity is lower in colonic polyps than in normal colon, and lower still in colonic neoplasia [62]. Reid and associates [63] observed factor XIIIa in both non-malignant and malignant fibroblastic and fibrohistiocytic lesions but reactivity was reduced in the malignant lesions. Roch et al. [64] showed that transglutaminase activity is readily detectable in the glandular epithelium of normal breast tissue but is usually not detectable in breast cancer or its metastases. In several cases in which weak transglutaminase activity was detected, localized disease was present. These results suggest that tissue transglutaminases do not likely contribute to stabilization of fibrinogen in the extracellular matrix of cancer tissues.

Conclusions Fibrinogen is a conspicuous component of the connective tissue stroma of many human tumor types. Fibrin, in contrast, has been demonstrated to be restricted in distribution to the tumor vasculature [38] or to tumor-associated macrophages in certain tumor types [34, 40], but such fibrin is unlikely to contribute to tumor progression. Fibrin associated with tumor cells that also express a procoagulant pathway is a feature of a minority of human tumor types [28-33] and indications exist that tumor cell-induced thrombin generation and fibrin formation may indeed contribute to tumor progression in these malignancies. This conclusion is based on the favorable results of therapeutic trials designed to limit thrombin activity and fibrin formation, or to remove tumor-associated fibrin [17-22].

Key unanswered questions - What is the composition of 'thrombi' that are

288

-

-

-

-

-

-

-

-

associated with tumor cells within blood vessels? What is the physico-chemical state of the fibrinogen that is present within the connective tissue stroma of many different tumor types, and particularly those tumor types lacking tumorassociated fibrin formation? Can such fibrinogen sequestration within tumors be accounted for by its ability to attach specifically to tumor cells or to other components of the extracellular matrix? Are there human tumor types other than the few that have been discovered so far that also manifest tumor cell procoagulants and tumor cell associated fibrin formation? What is the specific role of fibrin in the economy of those malignancies characterized by tumor fibrin formation? What is the initiator of the coagulation pathway on tumor cells in vivo that leads to fibrin formation? To what extent can the course of fibrin-containing human tumor types be improved by therapy with more effective thrombin inhibitors and fibrinolytic drugs than have been tested so far? Is it possible to improve on the therapeutic benefits of immune enhancers by testing anticoagulants given concomitantly in clinical trials as has been shown for experimental animal tumors? Considering the limited efficacy of current cancer therapies and the highly promising results obtained in certain pilot and prospective randomized clinical trials, why has this experimental approach to the management of malignancy not enjoyed a higher priority for testing?

A c k n o w l e d g e m e n t s

Supported in part by the Italian Association for Cancer Research (A.I.R.C.), Milan, Italy (to Vincenzo Costantini, M.D.) and the Department of Veterans Affairs Medical Research Service (Leo R. Zacharski, M.D.).

R e f e r e n c e s

1. Zacharski LR: Anticoagulation in the treatment of cancer in man. In: Donati MB, Davidson J, Garattini S (ed) Malignancy and the hemostatic system. Raven Press, New York, pp 113-128, 1981 2. Pickles FR, Edwards RL: Activation of blood coagulation in patients with cancer: Trousseau's syndrome revisited. Blood 62: 14-31, 1983 3. Donati MB, Poggi A, Semeraro N: Coagulation and malignancy. In: Poller L (ed) Recent advances in blood coagulation. Livingston, New York, pp 375-391, 1981 4. Zacharski LR, Henderson WG, Rickles FR, Forman WG, Cornell CJ, Harrower HW, Johnson RO: Rationale and experimental design for the VA Cooperative study of anticoagulation (warfarin) in the treatment of cancer. Cancer 44: 732-741, 1979 5. Rickles FR, Hancock WW, Edwards RL, Zacharski LR: Antimetastatic agents. I. The role of cellular procoagulants in the pathogenesis of fibrin deposition in cancer and the use of anticoagulants and/or antiplatelet drugs in cancer treatment. Sem Thrombos Hemostas 14: 88-94, 1988 6. Dvorak HF: Tumors: Wounds that do not heal. New Engl J Med 315: 1650-1659, 1986 7. Dvorak I-IF: Thrombosis and cancer. Human Pathol 18: 275-284, 1987 8. Dvorak HF, Harvey VS, Buchinski B, Estrella P, McDonagh J, Dvorak AM: Fibringels induce angiogenesis: a new assay with implications for tumor stroma generation and wound healing. Fed Proc 46: 970, 1987 9. Brown LF, Asch B, Harvey VS, Buchinski B, Dvorak HF: Fibrinogen influx and accumulations of cross-linked fibrin in mouse carcinomas. Cancer Res 48: 1920-1925, 1988 10. Dvorak HF, Scnger DR, Dvorak AM: Fibrin as a componant of the tumor stroma: origins and biological significance. Cancer Metast Rev 2: 41-73, 1983 11. Zacharski LR, Memoli VA, Costantini V, Wojtukiewicz MZ, Omstein DL: Clotting factors in tumor tissue: Implications for cancer therapy. Blood Coag Fibrinolys 1: 71-78, 1990 12. Zacharski LR, Howell AL, Memoli VA: The coagulation biology of cancer. Fibrinolysis 6 (Suppl): 39-42, 1992 13. Cardinali M, Uchino R, Chung SI: Interaction of fibrinogen with murine melanoma cells: covalent association with cell membranes and protection against recognition by lumpholkine-activated killer cells. Cancer Res 50: 80108016, 1990 14. Gunji Y, Gorelik E: Role of fibrincoagulation in protection of murine tumor cells from destruction by immune cells. Cancer Rcs 48: 5216-5221, 1988 15. Zacharski LR, Henderson WG, Rickles FR, Forman WG, Van Eeckhout JP, Cornell CJ, Forcier RJ, Martin JF: Rationale and experimental design for the VA Cooperative Study of RA-233 in the treatment of cancer. Am J Clin Oncol 5: 593--609,1982

289 16. Karpatkin S, Pearlstein E: Role of platelets in tumor cell metastasis. Ann Int Med 95: 636--641, 1981 17. Zacharski LR, Henderson WG, Rickles FR, Forman WB, Cornell CJ, Foreier RJ, Headley E, Kim S-H: Effect of sodium warfarin on survival in small cell carcinoma of the lung. J Am Med Assoc 245: 831-835, 1981 18. Zaeharski LR, Henderson WG, Rickles FR, Forman WB, Cornell CJ, Forcier RJ, Edwards RL, Headley E: Effect of warfarin anticoagulation on survival in carcinoma of the lung, colon, head and neck, and prostate. Final report of the VA Cooperative Study 75. Cancer 53: 2046-2052, 1984 19. Chahinian AP, Propert KJ, Ware JH, Zimmer B, Perry MC, Hirsh V, Skarin A, Kopel S: A randomized trial of antieoagulation with warfarin and of alternating chemotherapy in extensive small-cell lung cancer by the Cancer and Leukemia Group B. J Clin Oncol 7: 993--1002, 1989 20. Calvo PA, Harguindey SS, Aparicio LA, Dy C, Gil A, Rocha E: Urokinase and combination chemotherapy for treatment of small cell carcinoma of the lung. Cancer Treat Sympos 2: 105-108, 1985 21. Calvo FA, Santos M, Hidalgo OF: Urokinase-combination chemotherapy in small cell lung cancer: A phase II study. Proc ASCO 9: 237, 1991 22. Lebeau B, Chastang CL, Brechot JM: Subcutaneous hepafin treatment increases complete response rate and overall survival in small cell lung cancer (SCLC). Lung Cancer 7 (Suppl): 129, 1991 23. Donati MB, Davidson JF, Garattini S: Mafignancy and the hemostatic system. Raven Press, N.Y., 1981 24. Salsali M, Cliffton EE: Superior vena caval obstruction with lung cancer. Ann Thoracic Surg 6: 437--442, 1968 25. Astedt B, Glifberg I, Mattsson W, Trope C: Arrest of growth of ovarian tumor by tranexamic acid. J Am Med Assn 238: 154-155, 1977 26. Kudryk B, Rohoza A, Ahadi M, Chin J, Wiebe ME: Specificity of a monoclonal antibody for the NH2-terminal region of fibrin. Mol Immuno121: 89-94, 1984 27. Kudryk B, Grossman ZD, McAfee JG, Rosebrough SF: Monodonal antibodies as probes for fibrin(ogen) proteolysis. In: Chatal JF (ed) CRC Press, Boca Raton, Florida, pp 365-398, 1989 28. Zacharski LR, Memoli VA, Rousseau SM, Kisiel W: Coagulation-cancer interaction in situ in small cell carcinoma of the lung. Cancer 60: 2675-2681, 1987 29. Zacharski LR, Memoli VA, Rousseau SM" Thrombin-speeific sites of fibrinogen in small cell carcinoma of the lung. Cancer 62: 299-302, 1988 30. Wojtukiewicz MZ, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Rousseau SM, Stump DC: Abnormal regulation of coagulation/fibrinolysis in small cell carcinoma of the lung. Cancer 65: 481--485, 1990 31. Zacharski LR, Memoli VA, Rousseau SM: Cancer-coagulation interaction in situ in renal cell carcinoma. Blood 68: 394-399, 1986 32. Wojtukiewicz MZ, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Rousseau SM, Stump DC: Fibrinogen-fibrin

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

transformation in situ in renal cell carcinoma. Antieancer Res 10: 579-582, 1990 Wojtukiewicz MZ, Zaeharski LR, Memoli VA, Kisiel W, Kudryk BJ, Rouseau SA, Stump DC: Malignant melanoma: Interaction with coagulation and fibrinolysis pathways in situ. Am J Ciln Patho193: 516-521, 1990 Ornstein D, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Hunt J, Rousseau SM, Stump DC: Coexistant macrophage procoagnlant and tumor cell plasminogen activator in adenocarcinoma and squamous cell carcinoma of the lung. Cancer 68: 1061--67, 1991 Costantini V, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Hunt J, Rousseau SM, Stump DC: Fibrinogen deposition without thrombin generation in primary human breast cancer tissue. Cancer Res 51: 349-353, 1991 Costantini V, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Hunt J, Rousseau SM, Stump DC: Occurence of components of fibrinolysis pathways in situ in neoplastic and non-neoplastic human breast tissue. Cancer Res 51: 354-358, 1991 Wojtukiewicz MZ, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Rousseau SM, Stump DC: Indirect activation of blood coagulation in colon cancer. Thrombosis and Haemostasis 62: 1062-1066, 1989 WojtukiewiczMZ, Zacharski LR, Memoli VA, Moritz TE, Kisiel W, Kudryk BJ, Rousseau SM, Stump DC: Fibrin formation on vessel walls in hyperplastic and malignant prostate tissue. Cancer 67:1991 Wojtukiewicz MZ, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Rousseau SM, Stump DC: Absence of components of coagulation and fibrinolysis pathways in situ in mesothelioma. Thrombos Res 55: 279-284, 1989 Costantini V, Zacharski LR, Memoli VA, Kisiel W, Kudryk BJ, Rousseau SM, Stump DC: Fibrinogen deposition and macrophage-associated fibrin formation in malignant and non-malignant lymphoid tissue. J Lab Clin Med 119: 124-131, 1992 Zacharski LR, Dunwiddie C, Nutt EM, Hunt J, Memoli VA: Cellular localization of activated factor X by a Xaspecific probe. Thromb Haemost 65: 545-548, 1991 Werling RW, Zacharski LR, Kisiel W, Bajaj SP, Memoli VA, Rousseau SM: Distribution of tissue factor pathway inhibitor (TFPI, EPI, LACI) in normal and pathologic human tissues. Blood 78: 72a, 1991 Zacharski LR, Dunwiddie C, Nutt EM, Hunt J, Memoli VA: Cellular localization of activated factor X by a Xaspecific probe. Thromb Haemost 65: 545-548, 1991 Fingert HJ, Chen Z, Migraki N, Gajewaki WH, Bamberg MP, Kradin RL: Rapid growth of human cancer cells in a mouse model with fibrin clot subrenal capsule assay. Cancer Res 47: 3824-3829, 1987 Bogden AE, Cobb WR, Lepage D J, Haskell PM, Gulkin TA, Ward A, Kelton DE, Esber HJ: Chemotherapy responsiveness of human tumors as first transplant generation xenografts in the normal mouse: six day subrenal capsule assay. Cancer 48: 10--20, 1981

290 46. Olander JV, Bremer ME, Marasa JC, Feder J: Fibrinenhanced endothelial cell organization. J Cell Physiol 125: 1-9, 1985 47. Blomback B, Okada M, Forslind B, Larsson U: Fibrin gels as biological filters and interfaces. Biorheology 21: 93-104, 1984 48. Colvin RB, Gardner PI, Roblin RO, Verderber EL, Lanigan JM, Mosesson MW: Cell surface fibrinogen-fibrin receptors on cultured human fibroblasts. Lab Invest 41: 464473, 1979 49. Azzarone B, Frouty-Boye D, Macieira-Coelho A: Relationship between fibrin clot retraction and tumorigenesis in C3H/1071/2 cells. Int J Cancer 28: 799-803, 1981 50. Curatolo L, Azzarone B, Fally-Crepin C, Morasca L, Macieira-Coelho A: Comparison of fibrin clot retraction with other transformation parameters after hybridization of normal and established cell lines. Int J Cancer 31: 249-255, 1983 51. Hopper KE, Geczy CL, Davies WA: A mechanism of migration inhibition in delayed-type hypersensitivity reactions. I Fibrin deposition on the surface of elicited peritoneal macrophages in vivo. J Immunol 126: 1052-7, 1981 52. Sultan AM, Dunn CJ, Willoughby DA: Leukocyte migration inhibition activity of nonimmune acute inflammatory pleural exudate. Inflammation 3: 305-317, 1979 53. Ciano PS, Colvin RB, Dvorak AM, McDonagh J, Dvorak HF: Macrophage migration in fibrin gel matrices. Lab Invest 54: 62-70, 1986 54. Cheresh DA, Smith JW, Cooper HM, Quaranta V: A novel vitronectin receptor integrin (alpha V beta X) is responsible for distinct adhesive properties of carcinoma cells. Cell 57: 59-69, 1989 55. Cheresh DA, Siro RC: Biosynthetic and functional properties of an Arg-Gly-Asp-directed receptor involved in human melanoma cell attachment to vitronection, fibrinogen, and von Willebrand factor. J Biol Chem 262: 17703-17711, 1987 56. Ylanne J, Hormia M, Jarvinen M, Vartio T, Virtanen I: Platelet glycoprotein IIG IIIa complex in cultured cells. Localization in focal adhesion sites in spreading HEL cells. Blood 72: 1478-1486, 1988 57. Sosman JA, Kohler PC, Hank J, Moore KH, Bechhofer R,

58.

59.

60.

61.

62.

63.

64.

Storer B, Sondel PM: Repetitive weekly cycles of recombinant human interleukin-2: responses of renal carcinoma with acceptable toxicity. J Natl Cancer Inst 80: 60--63,1988 West WH, Tauer KW, Yannelli JR, Marshall GD, Orr DW, Thurmen GB, Oldham RK: Constant-infusion recombinant interleukin-2 in adoptive immunotherapy of advanced cancer. N Engl J Med 316: 898--905, 1987 Greenberg CS, Birckbichler PJ, Rice RH: Transglutaminases: multifunctional cross-linking enzymes that stabilize tissues. FASEB J 5: 3071-7, 1991 Murthy SNP, Lorand L: Cross-linked Ax.8 chain hybrids serve as unique markers for fibrinogen polymerized by tissue transglutaminase. Proc Natl Acad Sci 87: 9679-9682, 1990 Alaoui SE, Legastelois S, Roch AM, Chantepie J, Quash G: Transglutaminase activity and NE (y-glutamyl) lysine isopeptide levels during cell growth: an enzymic and immunological study. Int J Cancer 48: 221-226, 1991 Roch AM, Duchet M, Lointier P, Quash G: Polyamine biosynthesis, degradation and sequestration in tumors, polyps and histologically normal mucosa from the human colon. In: Perin A, Scalabrino G, Sessa A, Ferioli ME (ed) Perspectives in polyamine research. Wichtig, Milan, 129132, 1988 Reid MB, Gray C, Fear JD, Bird CC: Immunohistological demonstration of factors XIIIa and XIIIs in reactive and neoplastic fibroblastic and fibrohistiocytic lesions. Histopathol 10: 1171-1178, 1986 Roch AM, Noel P, Alaoini SE, Chariot C, Quash G: Differential expression of isopeptide bonds NE (y-glutamyl) lysine in benign and malignant human breast lesions: an immunohistochemical study. Int J Cancer 48: 215-220,1991

Address for offprints: Vincenzo Costantini, Institute of Internal and Vascular Medicine, University of Perugia, Via E.dal Pozzo, 1-06100Perugia, Italy

The role of fibrin in tumor metastasis.

A volume of data that has accumulated for over a century has suggested that fibrin may facilitate the persistence and progression of malignancy. Techn...
702KB Sizes 0 Downloads 0 Views