Angiogenesis and angiogenic mediators in haematological malignancies

British Journal of Haematology, 2000, 111, 43±51 Review ANGIOGENESIS AND ANGIOGENIC MEDIATORS IN HAEMATOLOGICAL MALIGNANCIES Neovascularization of solid tumours is suggested to be important for tumour growth and metastasis (Folkman et al, 1989). Various investigators have reported that tumours promote angiogenesis by secreting growth factors that stimulate endothelial cell migration and capillary proliferation (Folkman, 1995a,b; Mantovani et al, 1997; Nicosia, 1998) Indeed, angiogenic activity in a given tumour governs the potential for metastases and inhibition of angiogenesis may prove significant in suppressing neoplastic growth and invasiveness (Folkman, 1995b; Boehm et al, 1997; Cheresh, 1998; Benjamin et al, 1999). The role of angiogenesis in haematological malignancies is beginning to be identified (Vacca et al, 1995, 1997; Ribatti et al, 1996; Perez-Atayde et al, 1997). It is reported that the circulation within the bone marrow is similar to lymphoid tissues, i.e. lymph nodes, liver and spleen, and the lymphohaemopoietic tissue growth and function is dependent on its intravascular bed (Branemark, 1968; Bruyn et al, 1970). The prognostic role of neovascularization in lymphohaemopoietic malignancies remains contentious. This review focuses on the development of the microvascular bed, assessment of angiogenesis, its prognostic value and the possible therapeutic role of antiangiogeneic factors in haematological malignancies. Haemopoiesis, vasculogenesis and angiogenesis In the human yolk sac, vascular and haemopoietic tissues develop together by the development of yolk sac blood islands (Mangi & Layton, 1994; Palis et al, 1995). Angioblasts form the outer layer of blood islands, encasing multipotent haemopoietic cells (Fig 1). Angioblasts and yolk sac multipotent haemopoietic cells (haemocytoblasts) are CD34 positive (Mangi & Layton, 1994). The key gene instrumental in specifying the fate of multipotential mesodermal cells in the endothelial cell lineage has yet to be identified. It is suggested that vascular endothelial growth factors (VEGF, VEGF-B, VEGF-C, VEGF-D) and VEGF-related molecules (placental growth factors, PlGF) and their receptors [VEGFR-1 (flt-1), VEGFR-2 (KDR/flk-1), VEGFR-3] are important for normal development of blood vessels (Clark & Clark, 1939; Folkman & D'Amore, 1996; Cine et al, 1998; Nicosia, 1998; Darland & D'Amore, 1999), see Table I. Many investigators have defined vasculogenesis and angiogenesis as two different processes of vascular development that occur during early embryogenesis and postnatal Correspondence: Dr M. H. Mangi, Department of Haematology, The Royal London Hospital, Whitechapel, London E1 1BB, UK. E-mail: m.h.mangi@mds.qmw.ac.uk q 2000 Blackwell Science Ltd life (Folkman et al, 1996; Cine et al, 1998, Darland et al, 1999). Angiogenesis, which is the development of new capillaries from existing blood vessels, occurs in both the developing embryo and postnatal life (Asahara et al, 1997; Risau, 1997). Vasculogenesis involves the differentiation of mesodermal precursors to angioblasts that differentiate into endothelial cells to form the primitive capillary network (Risau & Flamme, 1995). It is suggested that vasculogenesis is limited to early embryogenesis and does not occur in adults (Risau et al, 1995; Cine et al, 1998; Yancopoulos et al, 1998). However, a recent report by Shi et al (1998) suggests that the process of vasculogenesis is not restricted to early embryogenesis and this may have physiological and pathological roles in health and disease in adults. A more complete understanding of the scope of vasculogenesis in humans may hold the promise of improved treatment strategies for vascular disorders in the future. Physiological and pathological angiogenesis The cellular and molecular mechanisms governing vasculogenesis and angiogenesis in experimental mouse embryos suggest that vascular endothelial growth factors and fibroblast growth factors are essential for initiation of vascular development (Clark et al, 1939; Darland et al, 1999). The formation and expansion of the vascular wall is further controlled by tie-1 and tie22 receptors, which are members of the RTK (receptor tyrosine kinase) family (Cine et al, 1998). Other factors that work in combination with VEGF are angiopoietin-1 and -2 that bind to tie-2. VEGF and angiopoietin-2 lead to angiogenesis and withdrawl of angiopoietin-2 can lead to the regression of capillaries (Sato et al, 1995). Angiopoietin-1 is mainly responsible for the maintenance of mature vessels and mutations of tie-2 or angiopoietin-1 may lead to embryos with inadequate vessel walls and abnormal hearts (Risau et al, 1995; Risau, 1997; Yancopoulos et al, 1998). Specific tie-2 mutations can be associated with smooth muscle deficiencies and microaneurysms (Vikkula et al, 1996). Although physiological neovasculogenesis is a self-limiting process, pathological angiogenesis persists over a longer period of time. This is seen in a variety of disorders such as rheumatoid arthritis, psoriasis, scleroderma, diabetic retinopathy and solid tumour growth (Koch et al, 1986; Folkman, 1995a; Koch, 1998). It is proposed that angiogenesis in humans is regulated by a delicate balance of proangiogenic and antiangiogenic factors. Proangiogenic factors include vascular endothelial growth factor (VEGF), acidic and basic fibroblast growth factors (aFGF, bFGF), angiogenin, angiopoietin-1, transforming growth factors (TGF-a, TGF-b), 43 44 Review Fig 1. Yolk sac haemopoiesis and endothelial development: endothelial cells encasing yolk sac blood islands (H & E stain, original magnification 40). tumour necrosis factor (TNF), platelet-derived growth factor (PDGF), platelet-derived endothelial cell growth factor (PDECGF), interleukin (IL)-2, IL-6, granulocyte colony-stimulating factor (G-CSF), granulocyte±macrophage (GM)-CSF, epidermal growth factor (EGF), insulin-like growth factor (IGF-1) and hepatocyte growth factor (HGF). There are naturally occurring and synthetic inhibitors of angiogenesis. Some commonly known antiangiogenic factors are located within larger proteins. These include the endostatin fragment of XVIII collagen, the platelet factor-4 fragment, the epidermal growth factor fragment, thrombospondin, fibronectin, prolactin, angiostatin plasminogen fragments, peptides of type 1 collagen and the tissue inhibitor of matrixmetalloproteinase (TIMP). Other inhibitors include IL-12, a-, b- and g-interferon (IFN), retinoic acid and thalidomide (see Table I). ANGIOGENESIS, INTEGRINS AND APOPTOSIS It is reported that cell adhesion molecules play an important role in angiogenesis. During new capillary development, coordinated signals from both integrins and growth factor receptors regulate the survival, proliferation and invasion of endothelial cells. To control increased angiogenesis, various Table I. Proangiogenic and antiangiogenic factors. Selected proangiogenic factors Selected antiangiogenic factors Vascular endothelial growth factor (VEGF) VEGF-B VEGF-C VEGF-D Placental growth factor (PIGF) Basic/acidic fibroblast growth factor (FGF) Angiogenin Angiopoietin-1 Platelet-derived growth factor (PDGF) Platelet-derived epidermal growth factor Hepatocyte growth factor (HGF) Epidermal growth factor (EGF) Insulin-like growth factor (IGF-1) Tumour necrosis factor a (TNF-a) Transforming growth factor (TGF-a and -b) Granulocyte±macrophage colony-stimulating factor (GM-CSF) Granulocyte colony-stimulating factor (G-CSF) Interleukin (IL)-2 IL-6 IL-8 Vitaxin aVb3 Endostatin Angiostatin g- and a-Interferon Thrombospondin Fibronectin Platelet factor 4 fragment Epidermal growth factor fragment Tissue inhibitor of metalloproteinases Retinoic acid Thalidomide IL-1 IL-12 Anti -VEGF Anti-vitaxin Anti-aVb3 q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 Review investigators have devised strategies to target growth factors or their receptors or integrins. Important integrins that control angiogenesis are the aVb3 and aVb5 integrins (Brooks et al, 1994a; Cheresh, 1998). Various preclinical studies have used antibodies to aVb3 to control increased angiogenesis, thereby limiting tumour growth. Brooks et al (1994b) reported that application of bFGF in combination with anti-aVb3 integrin resulted in inhibition of corneal angiogenesis. When the same experiments were repeated by incubating corneas with TNF-a and anti-aVb3 and aVb5, similar inhibition of angiogenesis was observed. These data suggest that bFGF and TNF-a activate endothelial cells via the aVb3 and aVb5 integrins. Further experiments showed that aVb3 and aVb5 integrins potentiate and promote endothelial cell survival signals by suppressing the activity of the p53 tumour-suppressor gene (Stromblad et al, 1996). The use of anti-aVb3, TNF-a and g-IFN indicate that this cocktail disrupts angiogenesis and leads to the demise of malignant melanoma cells (Ruegg et al, 1998). It will be important to know why both cytokines TNF-a and g-IFN are essential in controlling angiogenesis in melanoma and whether or not endothelial aVb3 is the only integrin affected by this approach. Preclinical and phase I trials are under way to answer these questions. Angiogenesis and lymphangiogenesis Although various vascular endothelial growth factors act on both vascular and lymphatic endothelial receptors, there is a lack of lymphatic proliferation in tumours. This may have several explanations. First, lymphatics may not have vascular endothelial growth factor receptor (VEGFR). However, the discovery of VEGFR-3 (a specific lymphatic endothelial receptor activated by VEGF-C and VEGF-D) does not support this interpretation (Fallowfield & Cook, 1990; Nicosia, 1998; Salven et al, 1998). A second possible explanation may be that VEGF-C and -D may induce chemotactic factor production by the lymphatic endothelium rather than lymphangiogenesis, which leads to lymphatic invasion and tumour metastasis. Indeed expression of VEGF-C and VEGF-D on human lymphoma cells supports this theory (Nicosia, 1998). Third, the paucity of lymphatics in human tumours may be due to high interstitial tumour pressure, leading to collapse of intratumour lymphatics. Finally, this may be due to technical reasons, i.e. lack of specific lymphatic endothelial markers. The recent development of specific probes for VEGFR-3 may be more informative in the assessment of lymphangiogenesis in human tumours (Valtola et al, 1999). It will be of interest to compare neovascularization with neolymphangiogenesis in tumours and to determine whether this correlates with aggressive behaviour in different tumours. Haematological malignancies and angiogenesis Many investigators have assessed the role of neovascularization in adult and childhood acute leukaemias, chronic leukaemias, myelodysplastic syndromes (MDS), lymphomas, myelomas and chronic lymphocytic leukaemia (CLL). Different methods and various endothelial markers have been used in various studies (see Table II). Perez-Atayde et al 45 (1997) analysed bone marrow biopsies in 40 children with newly diagnosed acute lymphoblastic leukaemia (ALL). Bone marrow biopsy (BMB) endothelial cells were immunostained with mouse monoclonal antibody to CD31, CD34 and rabbit polyclonal antibody to factor VIII-related antigen (FVIIIrAg). Micovessels (mv) were counted at a magnification of 200. This study showed increased microvessel density (MVD) in leukaemia patients compared with controls. MVD in leukaemia patients was 51 and control MVD was 6, P , 0.0001. In addition, urinary basic fibroblast growth factor (bFGF) was measured in 22 patients and was increased in all 22 children with ALL, however there was no statistically significant difference in bFGF levels between children with newly diagnosed ALL and those in complete remission. Although this study stresses the importance of assessing MVD in BMB of childhood ALL, other important morphological and biological parameters such as the French±American±British (FAB) classification, immunophenotyping data, proliferation index, cytogenetics, gene mutations and translocations were not compared with increased angiogenesis in ALL. According to this study, simple tests such as serial urinary bFGF measurement showed decreased levels of bFGF with decreased tumour burden, but this was not reliable enough to differentiate between newly diagnosed ALL and ALL in complete remission. It should be noted that bFGF is increased in many other pathological conditions, including infections and cell proliferation and sustained high levels of bFGF should be interpreted with caution (Brunner et al, 1993; Nguyen et al, 1994). Aguayo et al (1998) assessed angiogenesis in the bone marrow biopsies of 82 adult patients with various myeloid and lymphoid disorders. These were MDS (n ˆ 24), acute myeloid leukaemia (AML; n ˆ 14), ALL (n ˆ 7), CML (n ˆ 20) and chronic lymphoblastic leukaemia (CLL; n ˆ 17). Bone marrow biopsies were immunostained with anti-factor VIII-related (FVIII-r) antibody and MVD was counted by utilizing digitized images and computer programming. There was a significant increase in the number of blood vessels in cases with CML and MDS, however BMBs of patients with AML, ALL and CLL did not show statistically significant increases in MVD compared with the controls (n ˆ 17). Contrary to this, high urinary levels of b-FGF and vascular endothelial factor (VEGF), as well as high MVD, has been reported in the bone marrow of cases with CLL by Kini et al (1998). This study used anti-CD31, anti-CD34 and anti-FVIII-r antibody to assess bone marrow angiogenesis in 11 cases of diffuse and nodular CLL. MVD was assessed at 600 high power field (hpf). Cases with CLL showed MVD of 16´09/hpf, controls were 7´4/hpf, P ˆ 0.0005. Other biological factors, i.e cell proliferation, transformation, etc., and technical differences could account for discrepancies in the results reported by various investigators. A multivariate analysis on large numbers of patients with CLL may elucidate the mechanisms of increased angiogenesis in CLL. Vacca et al (1995) assessed bone marrow angiogenesis in cases with multiple myeloma and reported positive correlation between increased BM angiogenesis and multiple myeloma cell proliferation. Contrary to this, Rajkumar et al q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 46 Review Table II Angiogenesis in haematological malignancies. Disease q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 ALL ALL ALL ALL AML AML CML MDS MDS/CMML CLL CLL Myeloma Myeloma Myeloma B-NHL Mycosis Fungoides NHL Various cancers NHL Number of patients 40 17 6 7 14 99 20 24 16 11 17 16 36 51 88 57 82 56 160 Endothelial markers FVIIIrAg CD31, CD34 FVIIIrAg FVIIIrAg FVIIIrAg FVIIIrAg FVIIIrAg FVIIIrAg CD31, UEA-1 FVIIIrAg, CD31, CD34 FVIIIrAg FVIIIrAg CD34 FVIIIrAg FVIIIrAg, E/M FVIIIrAg MMP-2, MMP-9 NA NA NA Serum/urinary blood/platelet VEGF/sFGF/uFGF BMB-MVD Correlation with other prognostic/ biological features References " u-FGF NA "ubFGF NA NA "VEGF NA NA NA "ubFGF NA NA NA NA NA NA " " " " NS NA " " " " " " " " LN biopsy Skin biopsy NA NA NA NA NA "VEGF in high WCC AML(P0.01) NA NA "Extramedullary leukaemia NA NA NA Trisomy 13, CRP, b2M " cell proliferation " angiogenesis in high grade disease "angiogenesis in progressive disease Perez-Atayde et al (1997) Aguayo et al (1998) Veiga et al (1998) Aguayo et al (1998) Aguayo et al (1998) Aguayo et al (1999) Aguayo et al, (1998) Aguayo et al (1998) Mangi & Newland (1999) Kini et al (1998) Aguayo et al (1998) Rajkumar et al (1999) Munshi et al (1998) Vacca et al (1995) Ribatti et al (1996) Vacca et al (1997) Independent prognostic marker NA Independent prognostic marker Salven et al (1997) Salven et al (1999a) Salven et al (1999b) sVEGF" Lysed blood/platelets VEGF" sFGF" ALL, acute lymphoblastic leukaemia; AML, acute myeloid leukaemia; MDS, myelodysplastic syndrome; CMML, chronic myelomonocytic leukaemia; CLL, chronic lymphatic leukaemia; BMB, bone marrow biopsy; bFGF, basic fibroblast growth factor; u, urinary bFGF; s, serum bFGF; B, blood mononuclear cell bFGF; VEGF, vascular endothelial growth factor; MMP, matrixmetalloproteinases; NA, not available; WCC, white cell count; NS, not significant; LN, lymph node; ", increased. Review (1999) reported that increased angiogenesis was found in cases with multiple myeloma and that this phenomenon of increased capillary proliferation persisted even after stem cell transplantation. In addition, cases in complete remission also showed increased angiogenesis. The authors suggested that there may be a role for longterm anti-angiogenesis therapy in cases with myeloma. It is not clear from this study whether those patients who were receiving interferon showed similar or different patterns of angiogenesis. There are very few studies of angiogenesis in cases with lymphoma. Vacca et al (1997) and Ribatti et al (1996) reported increased capillary proliferation in the lymph node biopsies of high-grade non-Hodgkin's lymphoma (NHL) and skin biopsies of cases with progressive mycosis fungoides. Recent reports have assessed the concentration of basic fibroblast growth factor and VEGF in the sera of patients with high-grade NHL (Salven et al, 1997, 1999a,b). A total of 160 patients were analysed in this report, which indicated that high pretreatment levels of serum bFGF are associated with a poor prognosis in cases with large diffuse and immunoblastic lymphomas. A multivariate analysis suggested that a high serum bFGF level is an independent prognostic marker and it may provide more information about the course of the disease than the lactate dehydrogenase (LDH) levels and the number of extranodal sites in NHL (Salven et al, 1999b). A similar relationship between serum VEGF concentration and clinical indices in NHL has 47 been reported (Salven et al, 1997). These observations from a single centre are of considerable biological interest and larger studies are warranted to understand the pathophysiological and clinical relevance of serum VEGF/bFGF in haematologial malignancies. Most studies have not analysed any correlation between extramedullary leukaemic deposits and bone marrow angiogenesis. We assessed bone marrow angiogenesis in cases with myelodysplastic syndrome and chronic myelomonocytic leukaemia (CMML) with extramedullary leukaemic deposits using endothelial markers Ulex Europeus (UEA-1) and anti-CD31. A total of 16 cases were analysed in this study (Mangi & Newland, 1999). Two out of 16 cases with CMML and leukaemia cutis and 1 out of 16 cases with CMML and bladder chloroma showed the highest number of bone marrow and tissue microvessels (see Fig 2). This indicated that increased angiogenesis plays a role in extramedullary leukaemic deposits. Although the number of cases was too small to perform statistical analysis, our results suggest that antiangiogenic therapy may be considered in cases with extramedullary leukaemic deposits. Pitfalls of assessment of angiogenesis in haematological malignancies Some investigators have suggested that increased angiogenesis is of prognostic significance in solid tumours and haematological malignancies. However, many issues have not been addressed properly before these conclusions were Fig 2. Increased angiogenesis in a case with chronic myelomonocytic leukaemia with leukaemia cutis. Ulex europeus (UEA-1)-positive bone marrow trephine microvessels (original magnification 10). q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 48 Review reached. Basic facts may be correct but they can be interpreted in several different ways, e.g. substituting relative risk for absolute risk. We are aware of media hype and the `spin' that is put on some data and one has to take this into account when analysing evidence about increased angiogenesis in cases with haematological malignancies. The important points which require careful consideration are as follows. Endothelial labelling. Many techniques have been used to assess bone marrow biopsy angiogenesis. These techniques range from simple haematoxylin and eosin (H & E) stain to immunostaining using CD31, CD34, UEA-1 and FVIIIrelated antigen (Little et al, 1986; Mangi & Mufti, 1992). Polyclonal FVIIIr antigen can lead to some non-specific staining whereas CD34 stains endothelial cells and immature myeloid cells, which may lead to some overcounting of angiogenic hot-spots (Van de Rijin & Rouse, 1994). CD31 stains endothelial cells, platelets, megakaryocytes and some plasma cells (DeYoung et al, 1993; Weidner, 1995). In our experience, CD31 staining requires pretreatment with trypsin and it does not stain all endothelial cells. This may lead to undercounting of angiogenic areas in the bone marrow biopsy specimens. In addition, trypsin treatment to unmask CD31 in BMBs may not be the in vivo picture of angiogenesis. UEA-1 labels all endothelial cells, sinusoidal cells and megakaryocytes. This does not require trypsinization and probably is a more suitable marker to assess angiogenesis in BMB specimens (Little et al, 1986). It should be noted that UEA-1 does stain erythroid cells and stroma cells and, therefore, counting of angiogenic spots should be performed with caution. Computer compared with visual assessment. Brawer et al (1994) compared visual and automated counting of tumour microvessels by Optimas image analysis and found a good correlation (r2 ˆ 0.98, P , 0.001). In our experience, most endothelial markers stain other haemopoietic cells in bone marrow biopsies and computer-assisted assessment may overinterprete angiogenesis in cases with high numbers of megakaryocytes (UEA-1, CD31, FVIIIr Ag), bone marrow fibrosis (UEA-1, CD31), erythroid hyperplasia (UEA-1) and high numbers of myelomonocytic cells (CD34). We suggest that, in cases with myeloid leukaemia, both visual and computer-assisted analysis of microvessel hot-spots should be performed. Measurement of fibroblast growth factor (FGF) or VEGF. High levels of serum or urinary basic FGF and VEGF are suggestive of increased endothelial activity, infection, tissue breakdown and inflammation. High levels of bFGF have been demonstrated in cases with ALL and CLL. This may suggest high tumour burden or intercurrent infection or inflammatory process (Brunner et al, 1993). Comparison with other biological features. Many clinical and biological features in haematological malignancies are correlated with the clinical course of the disease. These include morphological features, immunophenotyping, cytogenetic abnormalities and translocations, tumour-suppressor genes, apoptosis, telomerase activity and in vitro culture studies. Most studies on angiogenesis have not performed multiparameter analysis to show an independent prognostic role for increased angiogenesis in haematological malignancies. Aguayo et al (1999) performed serum radioimmunoassay to detect levels of VEGF in AML patients with a high white cell count (WCC). This was a retrospective analysis and increased VEGF levels were compared with other parameters. This study suggests that increased VEGF is an independent prognostic marker (P ˆ 0.01) and has negative correlation with remission rates in AML patients with high WCC. The findings of increased angiogenesis in lymphohaemopoietic tumours should be interpreted with caution as most haematological cells produce or release angiogenic factors such as VEGF and bFGF. These include CD341 cells, monocytes, T cells, neutrophils, platelets and megakaryocytes (Koch et al, 1986; Brunner et al, 1993; Gaudry et al, 1997; Banks et al, 1999; Salven et al, 1999a). To what extent increased angiogenesis is physiological and merely reflects leucocyte numbers, which are clearly increased during most leukaemic disorders, remains contentious. Recently, Vermeulen et al (1999) compared high platelet count with high serum VEGF and bFGF levels in 58 cancer patients and reported that high bFGF levels were not associated with high leucocyte count or platelet count, whereas high serum VEGF levels had a significant association with high platelet count. A multivariate analysis entailing the lymphoma international prognostic index and serum bFGF was performed by Salven et al (1999b), who reported that high bFGF levels (. 5´5 pg/ml) were associated with poor survival in cases with high-grade NHL. Indeed, it has been proposed that serum bFGF provides more information than serum LDH levels and the number of extranodal tumour sites in NHL. Munshi et al (1998) showed correlation of increased angiogenesis with trisomy 13, C-reactive protein (CRP) and b2-microglobulin (b2M) in cases with multiple myeloma. We found a positive correlation between bone marrow angiogenic hot-spots and extramedullary leukaemic infiltrations in cases with CMML (Mangi & Newland, 1999). Further data on association between clinical indices, biological features and increased angiogenesis in haematological malignancies are warranted. Anti-angiogenesis therapy Many naturally occurring and synthetic inhibitors of angiogenesis are currently undergoing phase I and phase II trials. These include tissue inhibitors of metalloproteinase 1 and 2 (TIMP 1 and 2), anti-VEGF, inhibitor of plateletderived growth factor receptor, interferon a, IL-1, IL-12, TNF-a, retinoic acid, thalidomide, antibodies to integrin (vitaxin, aVb3), endostatin and angiostatin (O'Reilly et al, 1997; Koch, 1998). Preclinical studies in angiogenesis inhibitors (AIs) in animals have suggested that, by blocking the development of new blood vessels, the tumour can be destroyed (Boehm et al, 1997). Furthermore, this approach is less toxic, does not cause myelosuppression, hair loss and gastrointestinal symptoms and is not associated with cancer-associated drug resistance (Holmgren et al, 1995; Boehm et al, 1997; Kerbel, 1997). However, AIs have to be given daily or intermittently over a long period to achieve tumour control. This could be from months to years without q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 Review a break (Folkman, 1995b). Another approach is to use AIs along with cytotoxic drugs. This has led to the complete cure of tumours in some experimental animals (Teicher et al, 1994). Recently, Gordon et al (1998) evaluated safety and pharmacokinetics of recombinant human monoclonal antiVEGF antibody (anti-VEGF mAb) therapy in patients with various solid tumours. Twenty-five patients were entered in this phase I trial and received four doses of anti-VEGF mAb at a dose of 0´1±10 mg/kg intravenously (i.v.) over 90 min on days 0, 28, 35 and 42. It was concluded from this trial that anti-VEGF mAb is safe at doses up to 10 mg/kg. The follow-up period was too small to judge objective response in this trial, however one patient with renal cell tumour had a 39% reduction in tumour mass and three tumour-related bleeding episodes were seen. Further trials are under way to evaluate the effect of anti-angiogenesis therapy on overall survival and disease progression as well as tumour response. Recently, endostatin gained widespread attention after reports suggesting that a combination of endostatin and angiostatin eradicated Lewis lung tumour in mice (Boehm et al 1997). There were some difficulties in repeating these experiments by other investigators (Cohn, 1999). Further studies have been designed to evaluate the safety of endostatin in humans. The National Cancer Institute (NCI) has selected two centres to undertake a phase I trial of endostatin in patients with advanced solid tumours, including lymphomas, lung cancer, prostate cancer, colon cancer and breast cancer. With regard to the role of antiangiogenic therapy in other haematological malignancies, there is some interest, but it is difficult to find a group of patients who will clearly benefit from this therapy. There are some encouraging reports of thalidomide therapy in cases with relapsed myeloma (Singhal et al, 1999). One approach will be to use a combination of thalidomide and ainterferon therapy to achieve maximum response and improve overall survival in cases with myeloma. Other suitable patients may be cases with extramedullary leukaemia who clearly have angiogenic hot-spots in their bone marrow. The therapeutic implication of anti-angiogenesis therapy alone or in combination with cytotoxic therapy in these patients in preclinical or clinical settings is unknown. More studies on the role of proangiogenic and antiangiogenic markers in cases with myeloid and lymphoid malignancies are required to understand any role of increased angiogenesis, as well as lymphangiogenesis, in the progression of leukaemia and lymphoma. CONCLUSION Angiogenesis plays a key role in tumour growth, expansion and metastasis. This is well documented in solid tumours, including melanomas, ovarian cancer, lung cancer, colonic cancer, prostate cancer, brain tumours and some lymphomas. Assessment of angiogenic activity in haematological malignancies requires visual or computerized measurement of bone marrow microvessel density. This could be established with a variety of endothelial markers, including CD31, factor VIII-related antigen and Ulex Europeus (UEA-1). Many lines of evidence suggest increased 49 angiogenesis in cases with ALL, MDS, AML, multiple myeloma, NHL and CLL. There is a lack of reports on the correlation between increased angiogenesis with other biological data such as cytogenetic abnormalities and translocations, immunophenotypic data, oncogene activation and point mutations. Indeed, there are no consistent reports about the relationship between increased angiogenesis, remission rates, disease-free survival and overall survival. There are well-designed preclinical studies showing the effect of anti-angiogenesis therapy on retardation of tumour growth and cure of cancer. Obviously, these preclinical studies require further confirmation by human clinical trials. Nevertheless, a common concept is growing about angiogenesis, suggesting that future treatment of many tumours will require application of chemotherapy along with antiangiogenic therapy to improve overall survival in cancer. Department of Haematology, The Royal London Hospital, Whitechapel, London E1 1BB, UK Manzoor H. Mangi Adrian C. Newland REFERENCES Aguayo, A., Kantarjian, H., Talpaz, M., Estey, E., Estrov, Z., O'Brien, S., Keating, M., Barlogie, B. & Albitar, M. (1998) Increased angiogenesis in chronic myeloid leukemia and myelodysplastic syndromes (Abstract). Blood, 92, 607a. Aguayo, A., Estey, E., Kantarjian, H., Mansouri, T., Gidel, C., Keating, M., Giles, F., Estrov, Z., Barlogie, B. & Albitar, M. (1999) Cellular vascular endothelial growth factor is a predictor of outcome in patients with acute myeloid leukemia (AML). Blood, 94, 3717±3721. Asahara, T., Murohara, T., Sullivan, A., Silver, M., van der Zee, R., Li, T., Witzenbichler, B., Schatterman, G. & Isner, J.M. (1997) Isolation of putative progenitor endothelial cells for angiogenesis. Science, 275, 964±967. Banks, R.E., Forbes, M.A., Kinsey, S.E., Stanley, A., Ingham, E., Walters, C. & Selby, P.J. (1999) Release of the angiogenic cytokines VEGF from platelets. British Journal of Cancer, 79, 372±373. Benjamin, L.E., Golijanin, D., Itin, A., Pode, D. & Keshet, E. (1999) Selective ablation of immature blood vessels in established human tumors follows vascular endothelial growth factor withdrawl. Journal of Clinical Investigation, 103, 159±165. Boehm, T., Folkman, J., Browder, T. & O'Reilly, M.S. (1997) Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature, 390, 404±407. Branemark, P.-I. (1968) Bone marrow microvascular. Structure and function. Advances in Microcirculation, 1, 1±65. Brawer, M.K., Deering, R.E., Brown, M., Preston, S.D. & Bigler, S.A. (1994) Predictors of pathologic stage in prostate carcinoma. Cancer, 73, 678±687. Brooks, P.C., Clark, R.A.F. & Cheresh, D.A. (1994a) Requirement of vascular integrin aVb3 for angiogenesis. Science, 264, 569±571. Brooks, P.C., Montgomery, A.M., Rosenfeld, M., Reisfeld, R.A., Hu, T., Klier, G. & Chresh, D.A. (1994b) Integrin aVb3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell, 79, 1157±1164. Brunner, G., Nguyen, H., Gabrilove, J., Rifkin, D.B. & Wilson, E.L. q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 50 Review (1993) Basic fibroblast growth factor expression in human bone marrow and peripheral blood cells. Blood, 81, 631±638. Bruyn, P.P.H., Breen, P.C. & Thomas, T.B. (1970) The microcirculation of the bone marrow. Anatomy Records, 168, 55±68. Cheresh, D.A. (1998) Death to a blood vessel, death to a tumor. Nature Medicine, 4, 395±396. Cine, D.B., Pollak, E.S., Buck, C.A., Loscalzo, J., Zimmerman, G.A., McEver, R.P., Pober, J.S., Wick, T.M., Konkle, B.A., Schwatz, B.S., Barnathan, E.S., McCrae, K.R., Hug, B.A., Schmidt, A.M. & Stern, D.A. (1998) Endothelial cells in physiology and pathophysiology of vascular disorders. Blood, 91, 3527±3561. Clark, E.R. & Clark, E.L. (1939) Microscopic observation on the growth of blood capillaries in the living mammals. American Journal of Anatomy, 64, 251±299. Cohn, J. (1999) Behind the headlines of endostatin's ups and downs. Science, 283, 1250±1251. Darland, D.C. & D'Amore, P.A. (1999) Blood vessel maturation: vascular development comes of age. Journal of Clinical Investigation, 103, 157±158. DeYoung, B.R., Wick, M.R., Fitzgibbon, J.F., Sirgi, K.E. & Swanson, P.E. (1993) CD31: an immunospecific marker for endothelial differentiation in human neoplasms. Applied Immunohistochemistry, 1, 97±100. Fallowfield, M.E. & Cook, M.G. (1990) Lymphatics in primary cutaneous melanoma. American Journal of Surgical Pathology, 14, 370±374. Folkman, J. (1995a) Clinical applications of research on angiogenesis. New England Journal of Medicine, 333, 1757±1763. Folkman, J. (1995b) Angiogenesis in cancer, vascular rheumatoid and other diseases. Naure Medicine, 1, 27±31. Folkman, J. & D'Amore, P.A. (1996) Blood vessel formation. What is the molecular basis. Cell, 87, 1153±1155. Folkman, J., Watson, K., Ingber, D. & Hanahan, D. (1989) Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature, 339, 58±61. Gaudry, M., Bergerie, O., Andrieu, V., El Benna, J., Pocidalo, M.A. & Hakim, J. (1997) Intracellular pool of VEGF in human neutrophils. Blood, 90, 4153±4161. Gordon, M.S., Talpaz, K., Margolin, E., Holmgren, E., Sledge, Jr, G.W., Benjamin, R., Stalter, S., Shak, S. & Adelman, D. (1998) Phase I trial of recombinant humanized monoclonal antivascular endothelial growth factor (ANT-VEGF MAB) in patients with metastatic cancer (Abstract). Journal of Clinical Oncology, 17, 210a. Holmgren, L., O'Reilly, M.S. & Folkman, J. (1995) Dormancy of micrometastasis: balanced proliferation and apoptosis in the presence of angiogenesis suppresion. Nature Medicine, 1, 149± 153. Kerbel, R.S. (1997) A cancer therapy resistant to cancer. Nature, 390, 335±336. Kini, A.R., Peterson, L.C. & Kay, N.E. (1998) Evidence for abnormal angiogenesis in the bone marrow of patients with B-cell chronic lymphocytic leukemia (CLL) (Abstract). Blood, 92, 716a. Koch, A.E. (1998) Angiogenesis: implications for rheumatoid arthritis. Arthitis and Rheumatism, 41, 951±962. Koch, A.E., Polverini, P.J. & Leibovich, S.J. (1986) Induction of neovascularisation by activated human monocytes. Journal of Leukocyte Biology, 39, 233±238. Little, D., Said, J.W., Siegel, R.J., Fealy, M. & Fishbein, M.C. (1986) Endothelial cell markers in vascular neoplasms: an immunohistochemical study comparing FVIIIr Ag, blood group specific antigens, 6-keto PGF1 alpha and Ulex europeus 1 lectin. Journal of Pathology, 149, 89±95. Mangi, M.H. & Layton, D.M. (1994) The cells of human yolk sac blood island express CD34 and produce GM-CSF (Abstract). Blood, 84, 268a. Mangi, M.H. & Mufti, G.J. (1992) Primary myelodysplastic syndromes: diagnostic and prognostic significance of immunohistochemical assessment of bone marrow biopsies. Blood, 79, 198±205. Mangi, M.H. & Newland, A.C. (1999) Increased angiogenesis in chronic myelomonocytic leukaemia. Biologic and clinical implications. British Journal of Haematology, 105, 68a. Mantovani, A., Bussolino, F. & Introna, M. (1997) Cytokine regulation of endothelial cell function: from molecular level to bedside. Immunology Today, 18, 231±240. Munshi, N., Wilson, C.S., Penn, J., Epstein, J., Singhal, S., Hough, A., Sanderson, R., Desikan, R., Siegel, D., Mehta, J. & Barlogie, B. (1998) Angiogenesis in newly diagnosed multiple myeloma: poor prognosis with increased microvessel density in bone marrow biopsies (Abstract). Blood, 92, 98a. Nguyen, M., Watanable, I.I. & Budson, A.E. (1994) Elevated levels of an angiogenic peptide, basic fibroblast growth factor in the urine of patients with wide spectrum of cancers. Journal of the National Cancer Institute, 86, 356. Nicosia, R.F. (1998) What is the role of vascular endothelial growth factor-related molecules in tumor angiogenesis. American Journal of Pathology, 153, 11±16. O'Reilly, M.S., Boehm, T., Shing, Y., Fukai, N., Vasios, G., Lane, W.S., Flynn, E., Birkhead, J.R., Olsen, B.R. & Folkman, J. (1997) Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell, 88, 227±285. Palis, J., McGrath, K.E. & Kinsley, P.D. (1995) Initiation of hematopoiesis and vasculogenesis in murine yolk sac. Blood, 86, 156±163. Perez-Atayde, A.R., Sallan, S.E., Tedrow, U., Connors, S., Allred, E. & Folkman, J. (1997) Spectrum of tumor angiogenesis in the bone marrow of children with acute lymphoblastic leukemia. American Journal of Pathology, 150, 815±821. Rajkumar, S.V., Fonseca, R., Witzig, T.E., Gertz, M.A. & Greipp, P.R. (1999) Bone marrow angiogenesis in patients achieving complete response after stem cell transplantation for multiple myeloma. Leukemia, 13, 469±472. Ribatti, D., Vacca, A., Nico, B., Fanelli, M., Roncalli, L. & Dammacco, F. (1996) Angiogenesis spectrum in the stroma of B cell NHL. An immunohistochemical and ultrastructure study. European Journal of Haematology, 56, 45±53. Risau, W. (1997) Mechanism of angiogenesis. Nature, 386, 671± 674. Risau, W. & Flamme, I. (1995) Vasculogenesis. Annual Reviews of Cell Devision Biology, 11, 73±91. Ruegg, C., Yilmaz, A., Bieler, G., Bamat, J., Chaubert, P. & Lejeune, F.J. (1998) Evidence for the invovement of endothelial cell integrin Vb3 in the disruption of the tumor vasculature induced by tumor necrosis factor and interferon-gamma. Nature Medicine, 4, 408±414. Salven, P., Teerenhovi, L. & Joensuu, H. (1997) A high pretreatment serum vascular endothelial growth factor concentration is associated with poor outcome in non-Hodgkin's lymphoma. Blood, 90, 3167±3172. Salven, P., Lymboussaki, A., Jaaskela-Saari, H., Asae, K., von Euler, G., Eriksson, U., Alitalo, K. & Joesuu, H. (1998) Vascular endothelial growth factor VEGF-B and VEGF-C are expressed in human tumors. American Journal of Pathology, 153, 103±108. Salven, P., Orpana, A. & Joensuu, H. (1999a) Leukocytes and platelets of patients with cancer contain high levels of vascular endothelial growth factor. Clinical Cancer Research, 5, 487±491. Salven, P., Teerenhovi, L. & Joensuu, H. (1999b) A high q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51 Review pretreatment serum basic fibroblast growth factor concentration is an independent predictor of poor prognosis in non-Hodgkin's lymphoma. Blood, 94, 3334±3339. Sato, T.N., Tozawa, Y., Deutsch, U., Wolburg-Buchholz, K., Fujiwara, Y., Gendron-Maguire, M., Gridley, T., Wolburg, H., Risau, W. & Qin, Y. (1995) Distinct roles of the receptor tyrosine kinases Tie-1 and Tie-2 in blood vessel formation. Nature, 376, 70±74. Shi, Q., Rafi, S., Wu, M.H.D., Wijelath, E.S., Yu, C., Ishida, A., Fujita, Y., Kothari, R., Mohle, R., Sauvage, L.R., Moore, M.A.S., Storb, R.F. & Hammond, W.P. (1998) Evidence for circulating bone marrow derived endothelial cell. Blood, 92, 362±367. Singhal, S., Mehta, J., Desikan, R., Ayers, D., Roberson, P., Eddlemon, P., Munshi, N., Anaissie, E., Wilson, C., Dhodapkar, M., Zeldis, J. & Barlogie, B. (1999) Antitumor activity of thalidomide in refractory multiple myeloma. New England Journal of Medicine, 341, 1565±1571. Stromblad, S., Becker, J.C., Yerba, M., Brooks, P.C. & Cheresh, D.A. (1996) Suppression of p53 and p21 expression by vascular cell integrin aVb3 during angiogenesis in vivo. Journal of Clinical Investigation, 98, 426±433. Teicher, B.A., Holden, S.A., Ara, G., Sotomayor, E.A., Huang, Z.D., Chen, Y.-N. & Brem, H. (1994) Potentiation of cytotoxic cancer therapies by TNP-470 alone and with other antiangiogenic agents. International Journal of Cancer, 57, 920±925. Vacca, A., Di Loreto, M., Di Stefano, R., Gadaleta-Caldarola, G., Iodice, G., Caloro, D. & Dammacco, F. (1995) Bone marrow of patients with multiple myeloma: angiogenesis and plasma cell adhesion molecules LFA-1, VLA-4, LAM-1 and CD44. American Journal of Hematology, 50, 9±14. Vacca, A., Morretti, S., Ribatti, D., Pellegrino, A., Piminelli, N., Bianchi, B., Bonifazi, E., Ria, R., Serio, G. & Dammacco, F. (1997) 51 Progression of mycosis fungoides is associated with changes in angiogenesis and expression of matrix metalloproteinases 2 and 9. European Journal of Cancer, 33, 1685±1692. Valtola, R., Salven, P., Heikkila, P., Taipale, J., Joensuu, H., Rehn, M., Pihlajaniemi, T., Weich, H., deWaal, R. & Alitalo, K. (1999) VEGFR-3 and its ligand VEGF-C are associated with angiogenesis in breast cancer. American Journal of Pathology, 154, 1381±1390. Van de Rijin, M. & Rouse, R.V. (1994) CD34 a review. Applied Immunohistochemistry, 2, 71±80. Veiga, J.P., Hernani, A., Sallan, S.E., Nadler, L.M. & Cardoso, A.A. (1998) Acute lymphoblastic leukemia cells and their bone marrow stroma collaborate to induce endothelial cell proliferation thereby promoting leukemia cell survival (Abstract). Blood, 92, 511a. Vermeulen, P.B., Salven, P., Benoy, I., Gasparini, G. & Dirix, L.Y. (1999) Blood platelets and serum VEGF in cancer patients. British Journal of Cancer, 79, 370±371. Vikkula, M., Boou, L., Carraway, K., Calvert, J., Diamond, A., Gooumnerov, B., Pasyk, K., Marchunk, D., Warman, M., Cantley, L., Mulliken, J. & Oslen, B.R. (1996) Vascular dysmorphogenesis caused by an activating mutation in the receptor tyrosine kinase. Cell, 87, 1181±1191. Weidner, N. (1995) Intratumor microvessel density as a prognostic factor in cancer. American Journal of Pathology, 147, 9±19. Yancopoulos, G.D., Klagsbrun, M. & Folkman, J. (1998) Vasculogenesis, angiogenesis and growth factors: ephrins enter the fray at the border. Cell, 93, 661±664. Keywords: angiogenesis, vascular endothelial growth factors (VEGF), leukaemia, lymphoma, myeloma. q 2000 Blackwell Science Ltd, British Journal of Haematology 111: 43±51