The Emerging Role of Angiogenesis Inhibitors in Hematologic Malignancies

ABSTRACT: Angiogenesis is an important component of the pathogenesis of hematologic malignancies. A negative prognostic implication of increased angiogenesis has been established for acute and chronic myeloid and lymphocytic leukemias, myeloproliferative diseases, multiple myeloma, non-Hodgkin’s lymphoma (NHL), and hairy cell leukemia. An association between the return of increased marrow vascularity to normal levels and durability of response has been established in some of these diseases. Elevated levels of proangiogenic factors have been associated with a poor prognosis in the acute and chronic leukemias, multiple myeloma, and NHL. These data lend support to the reduction of activity of proangiogenic factors as a therapeutic modality. Vascular endothelial growth factor (VEGF) has been implicated as the major proangiogenic factor that regulates multiple endothelial cell functions, including mitogenesis. A direct relationship between VEGF and leukemic blasts and malignant plasma cells has been established, but VEGF may have a function distinct from its role in angiogenesis. Current protocols with anti-VEGF agents in patients with hematologic malignancies involve the use of monoclonal antibody, blockers of the VEGF-receptor tyrosine kinase pathway, thalidomide (Thalomid) and its analogs, and cyclooxygenase inhibitors. The receptor tyrosine kinase inhibitors also affect platelet-derived growth factor, c-kit, and Flt-3 to varying degrees, considerably broadening their potential efficacy. This review will summarize several angiogenesis inhibitors in clinical development. [ONCOLOGY 16(Suppl 4):23-29, 2002]

Data on the significant role ofangiogenesis in the hematologic disorders—including acute myeloid leukemia(AML), chronic myeloid leukemia (CML), acute lymphocytic leukemia (ALL), chroniclymphocytic leukemia (CLL), the myelodysplastic syndromes (MDS), non-Hodgkin’slymphomas (NHL), hairy cell leukemia, multiple myeloma, and agnogenic myeloidmetaplasia—is growing in an exponential manner (Table1).[1-26] Bone marrowmicrovessel density (MVD) is significantly greater in patients with advanced MDS(ie, refractory anemia with excess blasts [RAEB] or RAEB in transformation[RAEB-t]) compared with normal individuals. Patients with AML also have agreater bone marrow MVD,[10] and successful induction chemotherapy has resultedin a significant decrease in bone marrow MVD in these patients.[3] In childrenwith ALL, a complete remission (CR) associated with a return to normal bonemarrow MVD is more durable than a CR associated with a persistent increase inMVD.[4] In patients with AML, ALL, and MDS, in addition to the relative increasein MVD, there is a shift from the normal bone marrow predominance of straight,nonbranching microvessels to vessels with a complex, arborizing architecture. Inpatients with multiple myeloma, increased MVD has been directly correlated withthe plasma cell labeling index.[27]

The Proper Angiogenesis Modulator to Target

As summarized in Table 2, there are numerous molecules to targetwithin the angiogenesis cascade. However, one factor, vascular endothelialgrowth factor (VEGF), has emerged as the prime target in treating hematologicmalignancies. Vascular endothelial growth factor is pivotal in the angiogenicprocess and regulates several endothelial cell functions, including mitogenesis,permeability, vascular tone, and the production of vasoactive molecules.[28]Transcription of the VEGF gene (located on the short arm of chromosome 6 at6p21.3) is physiologically regulated by hypoxia. Production of VEGF falls underthe control of alternate mRNA splicing and proteolytic processing. Alternatesplicing is responsible for the production of the isoforms VEGF 189, VEGF 165,VEGF 121, and VEGF 205.[16]

The activity of VEGF is stimulated by three receptor tyrosinekinases: VEGFR-1 (Flt-1), VEGFR-2 (KDR/Flk-1), and VEGFR-3 (Flt-4). Primarilyexpressed on endothelial cells and monocytes, VEGFR-1 mediates cellmotility.[28] The proliferative and mitogenic activities of VEGF and increasesin vascular permeability are mediated primarily via VEGFR-2.[29] Finally,VEGFR-3, homologous to the neurophilin-1-receptor, is involved inlymphoangiogenesis.[30]

VEGF Activity in Hematologic Malignancies

Significantly elevated levels of VEGF are observed in a varietyof hematologic malignancies (Table 3). In a study of 417 patients, a number ofpatients with AML (n = 115) and advanced MDS (n = 40) had similar elevatedplasma levels of VEGF.[31] In stored serum samples, increasing levels ofcellular VEGF had a positive correlation with a lower CR rate, shorter CRduration, and shorter overall and disease-free survival times.[9] These data arein contrast to those reported for the receptor tyrosine kinase Tie-1, in whichwe used Western blot analysis to confirm and radioimmunoassay to quantify Tie-1protein expression in bone marrow samples obtained from untreated patients withAML (n = 66) or MDS (n = 29).[11] Tie-1 protein levels were elevated in alldisease specimens and were significantly higher in patients with AML than inpatients with MDS. However, Tie-1 levels did not correlate with CR or survivalduration in patients with either AML or MDS.

This contrast with respect to VEGF results is particularlyinteresting when one considers the relative roles of Tie-1 and VEGF in marrowangiogenesis. Tie-1 is expressed on both vascular endothelial cells and immaturehematopoietic cells, as is VEGF. Although Tie-1-deficient mice with a targetedinsertional mutation in their germ line have defects in endothelial cellintegrity, resulting in edema and hemorrhage, data indicate that Tie-1 is notcritical for either bone marrow stem cell engraftment or self-renewal.[32]Interactions between proangiogenic molecules and marrow stromal elements mayalso contribute to AML growth (Figure 1). For example, incubation of humanumbilical vein endothelial cells with VEGF results in increased secretion ofgranulocyte-macrophage colony-stimulating factor (GM-CSF), an AML blast growthfactor, by the endothelium.[33]

We have recently evaluated the clinical significance of VEGFR-1and VEGFR-2 protein levels in patients with untreated AML and MDS to identifyany relationship between these receptors and VEGF levels in each disease. Levelsof VEGFR-1 were significantly higher in AML than in MDS patients, and VEGFR-2levels were equivalent. There was no correlation between VEGFR levels andsurvival. Levels of VEGF were significantly higher in MDS than in AML patients,and levels correlated with poorer survival in MDS patients. There was asignificant correlation between VEGF and VEGFR-2 levels in both AML and MDSpatients, but not between VEGF and VEGFR-1 levels in either disease. These datasuggest that VEGF expression, not VEGFR expression, is of prognostic relevancein AML and MDS. The significance of the differential expression of VEGFR in AMLand MDS is unclear.

Chronic Lymphocytic Leukemia

Intracellular levels of VEGF have also been correlated withprognosis in patients with CLL.[1] In an analysis of samples collected frompatients (n = 225) with B-cell CLL, the median intracellular VEGF level was 7.26times higher than that of normal peripheral blood mononuclear cells. Patientswith lower levels of VEGF protein showed a trend toward a relatively reducedrate of survival. In a subgroup of patients with Rai stages 0 to II, Binet stageA and B disease, and good prognostic features (eg, beta-2-microglobulin of 2.8mg/dL or less), lower levels of VEGF were also associated with shorter survival times.However, no overall correlation among VEGF, disease stage, and beta-2-microglobulinlevels was demonstrated in the cohort.

In a separate report, there was an indirect correlation betweenelevations in serum VEGF and the duration of progression-free survival (PFS) inpatients with Binet stage A disease.[15] Molica et al combined the patients’Rai classification with serum VEGF levels, and identified two groups withdifferent PFS within stages I and II.[15] Patients with Rai stage I to II withelevated VEGF levels had very short PFS (median, 9.5 months) compared withpatients with Rai stage I to II with low VEGF levels, who had longer PFS (mediannot reached at 15 months).

In a recent examination of the role of VEGFR-2 in CLL, patientswith elevated VEGFR-2 levels had greater elevations in lymphocyte counts, severeanemia, elevated serum beta-2-microglobulin levels, and advanced-stagedisease.[19] Elevated VEGFR-2 levels were associated with reduced survival.Therefore, further studies on the interactive effects of VEGF and VEGFR-2 on CLLproliferation are warranted. An analysis of Flt-1 and Tie-1 protein levels inthis same cohort of CLL patients also showed that these protein levels correlatewith white blood cell counts and absolute lymphocyte counts.[34] Flt-1 proteinlevels correlated with levels of cellular VEGF, whereas levels of Tie-1 did not.Furthermore, neither correlated with survival in the overall cohort; however,higher levels of Tie-1, but not Flt-1, correlated with reduced survival.

Although high levels of both VEGF and basic fibroblast growthfactor (bFGF) have been documented in patients with CLL, bone marrow MVD is notincreased in most of these patients.[31] As observed in AML with VEGF levels,aside from any proangiogenic role, VEGF and bFGF have significant, independentroles in the pathophysiology of CLL.

Lymphomas

Elevated levels of VEGF are also associated with an adverseprognosis in patients with NHL.[25] However, the highest prognostic power wasobserved when VEGF and serum bFGF levels were examined in combination. The riskof death in patients whose VEGF and bFGF levels were both within the highestquartiles was greater than in other patients, independent of the prognosticvariables in the International Prognostic Index.[25] In collaboration with theOmaha group, we have recently analyzed the clinical significance of serum levelsof VEGF, bFGF, hepatocyte growth factor (HGF), and angiogenin in untreatedpatients with NHL or Hodgkin’s disease (HD). In patients with HD or NHL, VEGFand HGF levels were significantly increased (personal communication, M. Albitar,2002). In contrast, angiogenin levels were significantly decreased in patientswith either NHL or HD.

As reported by Salven et al, higher levels of VEGF and bFGF inNHL patients correlated with more advanced disease stage and with higher lactatedehydrogenase levels.[25] In addition, elevated levels of VEGF correlated withshorter survival in patients with HD. Elevated baseline VEGF levels tended toreturn to normal in patients with NHL who responded to therapy, and nosignificant changes in the levels of other factors were observed. In posttherapysamples collected from patients with HD, HGF and bFGF returned to normal levels,whereas VEGF levels remained elevated. Patients with NHL with higher posttherapyVEGF levels had relatively reduced survival and VEGF levels after therapyremained predictive of survival in patients with HD. Angiogenic factors appearto have distinct roles in the biology of HD and NHL, and a better appreciationof these differences may aid in the diagnosis of some patients.

Multiple Myeloma

In patients with multiple myeloma, levels of VEGF, bFGF, and HGFparallel disease activity, and VEGF levels correlate with features of aggressivedisease, including levels of serum C-reactive protein and beta-2-microglobulinl.[7] Multiple myeloma patients have significantly higher levels of VEGF in bonemarrow than in peripheral blood—malignant plasma cells have been shown toexpress and secrete VEGF. Although marrow MVD parallels disease activity in MM,and it is thus reasonable to postulate that VEGF is acting in an autocrinefashion, multiple myeloma cells have a low level of VEGFR expression. Thus, VEGFmay act in a paracrine fashion in multiple myeloma—both by its interactionswith interleukin-6, a known myeloma growth factor, and with bFGF. Levels of bFGFhave been reported to correlate with response rates to thalidomide (Thalomid)treatment in multiple myeloma patients.[35] Because HGF is overproduced inmultiple myeloma and malignant cells express the HGF receptor c-met, this may bethe basis for another autocrine loop in multiple myeloma.[36]

Cyclooxygenase and Hematologic Malignancies

Cyclooxygenase (COX, prostaglandin G/H synthase) is amembrane-bound enzyme responsible for the oxidation of arachidonic acid toprostaglandins, and this enzyme has two isoforms, COX-1 and COX-2.[37]Cyclooxygenase-1 is constitutively expressed and regulates homeostatic functionsincluding vascular hemostasis and gastroprotection. Cyclooxygenase-2 isinducible by mediators such as growth factors, cytokines, and endotoxins.Nonsteroidal anti-inflammatory drugs produce their therapeutic effects throughinhibition of COX prostaglandin formation. The expression of proangiogenicfactors, including VEGF, bFGF, transforming growth factor-beta, andinterleukin-6, is stimulated by prostaglandins.[38] Recent efforts have focusedon the development of selective inhibitors of COX-2, which avoid the adverseeffects associated with COX-1 inhibition. Many human solid tumors overexpressCOX-2 but not COX-1, and gene-knockout and transfection experiments demonstratea central role of COX-2 in experimental tumorigenesis.[39] Elevated COX-2 levelshave adverse prognostic significance and/or are associated with biologicfeatures favoring metastatic spread, recurrence, or local invasiveness in anumber of human solid tumors.

Cyclooxygenase inhibitors have been shown to decrease the rateof development of human colorectal tumors.[40] Cyclooxygenase-2 has also beenshown to affect the relationship between malignant cells and adjacent stroma.Abnormalities in this relationship abound in the bone marrow of patients withhematologic malignancies. Cyclooxygenase-2 has also been shown to regulatedifferential expression of apoptosis-related genes, thus favoring cell survivaland potentially mediating resistance to anticancer therapy.

COX-2 and UPP Inhibition

Perturbation of the apoptotic cascade is well documented inleukemia, multiple myeloma, and NHL, especially in terms of abnormal function ofmembers of the Bcl-2/Bax family. Selective inhibition of COX-2 has been shown toenhance cytotoxin-induced apoptosis in cancer models. We have recently observedthat COX-2 levels are significantly elevated in patients with CML. Nosignificant differences were evident in the levels of COX-2 expression inpatients with either early or late chronic phase CML. The elevation of COX-2 wasprognostically adverse in patients with chronic phase CML and was a moreimportant independent prognostic factor than the established CML staging systemsfor these patients. Cyclooxygenase-2 inhibition may be an important area ofinvestigation in hematologic malignancies, particularly in CML, AML, and MDS.

Another important pathway under investigation to indirectlysuppress VEGF involves inhibition of the ubiquitin-proteasome pathway (UPP). The26S proteasome, an adenosine triphosphate-dependent multicatalytic protease, isresponsible for the UPP breakdown of many intracellular proteins in eukaryoticcells.[41] The 26S proteasome (a core 20S particle bound to two regulatory 19Sparticles) lyses misfolded, oxidized, or damaged ubiquinated proteins, includingthose that regulate cell cycle and trafficking, transcription factor activation,and apoptosis. Cell mitosis is dependent on the orderly degradation of keyregulatory proteins by the UPP, including p53, cyclins, and the cyclin-dependentkinase inhibitors p21 and p27 K1p1.

PS-341 (aka LDP-341 or MLN-341) is a potent, sensitive,selective, and reversible small-molecule proteasome inhibitor that binds tightlyto the enzyme’s active sites.[42] The percentage inhibition of 20S proteasomeactivity in human blood cells is a reliable measure of the biologic activity ofPS-341. Significant responses to PS-341 have been observed in patients withhematologic malignancies, including in those with advanced refractory multiplemyeloma and AML. The antineoplastic effect of PS-341 appears to involve severalmechanisms, including the inhibition of cell growth signaling pathways andcellular adhesion molecule expression, the induction or facilitation ofapoptosis, and an antiangiogenic action possibly mediated by altering stromalcell-malignant cell interactions, with a consequent decrease in VEGF production.

Targeting Angiogenesis Modulators

If one focuses on VEGF, several obvious points of interventionemerge: decreasing production (antisense strategy or ribozymes), blockingreceptor binding (anti-VEGF monoclonal antibodies), and blocking signaltransduction. Indirect methods of VEGF suppression include COX-2 inhibition, UPPinhibition, and methods mediated by compounds such as aplidine (APL), in whichthe mechanism of action is not fully elucidated.

Angiogenesis Inhibitors

Bevacizumab (Avastin) is an anti-VEGF monoclonal antibodycomposed of a humanized murine antibody with antigen binding complementarydetermining regions from murine VEGF A.4.6.1.[43,44] It recognizes all VEGFisoforms but does not bind to bFGF, HGF, or platelet-derived growth factor(PDGF). It has potent indirect antitumor activity in experimental models. In aphase I, dose-escalation study involving 25 patients with refractory solidtumors, bevacizumab (0.1 to 10 mg/kg) was administered as a 90-minute infusionon days 0, 28, 35, and 42. Dose-limiting toxicities were not observed at weeklydoses of  10 mg/kg or less, although some patients experienced mild to moderateasthenia, mild headache, fatigue, nausea, and low-grade fever on the day of drugadministration. In addition, bleeding at tumor sites developed in three (12%)patients.

No data on single-agent bevacizumab in patients with hematologicmalignancies have been published; a study in MDS patients is ongoing. A NationalCancer Institute-sponsored phase II study in patients with blast phase CML, inwhich bevacizumab is given both prior to and with idarubicin (Idamycin) andcytarabine, is ongoing. In addition, a planned study will examine thecombination of bevacizumab and thalidomide in patients with refractory multiplemyeloma.

A number of VEGF-receptor kinase inhibitors are in clinicaldevelopment, including SU5416, SU6668, SU11248, PTK787/ZK 222584, ZD6474, andCGP41251, of which SU5416 is the furthest in clinical development for patientswith hematologic malignancies. SU5416 binds to the adenosine triphosphatebinding site of VEGFR-1 and VEGFR-2.[45,46] In addition, it also binds to PDGFreceptor and c-kit, a growth factor that promotes the survival of earlyhematopoietic progenitor cells and acts synergistically with other hematopoieticgrowth factors across multiple lineages.[47] The inhibition of c-kit is anemerging important therapeutic target in patients with AML.[48]

Another receptor tyrosine kinase of increasing interest fortreating AML is the Flt-3, which is expressed on leukemic blasts and mediatessurvival and proliferation. Internal tandem duplications in the Flt-3-juxtamembraneregion, which result in constitutive kinase activity, are found in approximately30% of AML patients, and their presence is associated with a poor prognosis.SU5416 inhibits phosphorylation of internal tandem duplication mutant Flt-3 bothin vitro and in xenograft mouse models.

SU5416 produces a dose-dependent inhibition of tumor growth in avariety of xenograft models. In a model of AML, SU5416 inhibited the stem cellfactor-induced proliferation of MO7e cells.[49] Incubation of MO7e cells withSU5416 induces apoptosis through activation of caspase-3 and increased poly(ADP-ribose) polymerase cleavage. These data were reproducible in blasts frompatients with AML. A phase I study established that the maximum tolerated doseof SU5416 was exceeded on a regimen of 190 mg/m²/d intravenously twice weeklyfor 4 weeks. The dose-limiting toxicities were projectile vomiting, headache,and nausea, all of which were reversible over a 24- to 48-hour period. Mild tomoderate toxicities included headache, pain at the infusion site, phlebitis,change in voice, and elevated aminotransferase levels.

Ongoing studies in patients with hematologic malignancies arebeing conducted at a dose of 145 mg/m² twice weekly on a 4-week cycle. In apilot study of SU5416 in 14 evaluable patients with c-kit-positive AML, majortoxicities included 1 fatal gastrointestinal bleed, 1 grade 4 pancreatitis, 1grade 4 hepatic transaminitis, and grade 2/3 bone pain in 3 patients.[50] Fivepatients demonstrated a partial response; however, 9 patients failed to respond,including 6 who had progressive disease during the study.

A US, multicenter, phase II trial has recently completedenrollment. Patients with refractory AML, MDS, CML, agnogenic myeloidmetaplasia, and multiple myeloma were treated with SU5416 145 mg/m² twice weeklyon a 4-week cycle. Objective responses have been reported in patients with AMLand agnogenic myeloid metaplasia in preliminary analyses of the results. SU6668,another small-molecule inhibitor of angiogenesis, is not being clinicallydeveloped for patients with hematologic malignancies. Another agent, SU11248, isa third-generation, orally available, VEGF- and PDGF-receptor tyrosinekinase inhibitor due to enter phase I/II studies in 2002 in patients withrefractory hematologic malignancies.

The agent PTK787/ZK222584 is also a potent and relativelyselective inhibitor of VEGFR-1 and VEGFR-2 and is active in the submicromolarrange.[51] At higher concentrations, it also inhibits other class III-receptortyrosine kinase inhibitors, including PDGF, c-kit, and c-fms. However, PTK787 isnot active against receptor tyrosine kinase from other families or intracellularkinases. It inhibits VEGF-induced endothelial cell proliferation, migration, andsurvival in the nanomolar range in cell-based assays. However, PTK787 does nothave cytotoxic or antiproliferative effects on cells that do not express VEGFR.The VEGF- and PDGF-induced angiogenesis are inhibited in a dose-dependent mannerby PTK787 in both a growth factor implant model and a tumor cell-drivenangiogenesis model after once-daily oral dosing (25 to 100 mg/kg). In the samedose range, PTK787 also inhibits the growth of several human carcinomas inanimal models by inhibiting microvessel formation in the interior of the tumor.However, PTK787 does not impair wound healing and has no overt effects oncirculating blood cells or bone marrow leukocyte levels.

In ongoing phase I studies, patients have been treated at anoral dose range of 50 mg to 2,000 mg once daily and 150 mg twice daily on acontinuous dosing schedule. In addition, phase I/II studies in patients withrefractory hematologic malignancies are due to begin in 2002.

Conclusions

It is immediately apparent that one or more of the angiogenesisinhibitors exhibit sufficient clinical activity in patients with hematologicmalignancies to stimulate large-scale commercial interest in further clinicaldevelopment. The traditional bias against development of drugs in these uncommonhematologic disorders (in favor of development in solid tumors) must be resisted—invitro evidence that angiogenesis is important in leukemia is more convincingthan that for any other human malignancy. Novel clinical trial designs areclearly needed. The multiple signaling pathways activated by VEGF, includingmitogen-activated protein kinase, focal adhesion kinase, phosphatidylinositol3-kinase, protein kinase B, protein kinase C, and paxillin, are each beingindependently investigated as potential therapeutic targets. In addition,combinations of VEGF inhibitors with PS-341 appear particularly promising.

References:

1. Aguayo A, O’Brien S, Keating M, et al: Clinical relevanceof intracellular vascular endothelial growth factor levels in B-cell chroniclymphocytic leukemia. Blood 96:768-770, 2000.

2. Cao Y, Veitonmaki N, Keough K, et al: Elevated levels ofurine angiostatin and plasminogen/plasmin in cancer patients. Int J Mol Med5:547-551, 2000.

3. Padro T, Ruiz S, Bieker R, et al: Increased angiogenesis inthe bone marrow of patients with acute myeloid leukemia. Blood 95:2637-2644,2000.

4. Perez-Atayde AR, Sallan SE, Tedrow U, et al: Spectrum oftumor angiogenesis in the bone marrow of children with acute lymphoblasticleukemia. Am J Pathol 150:815-821, 1997.

5. Hussong JW, Rodgers GM, Shami PJ: Evidence of increasedangiogenesis in patients with acute myeloid leukemia. Blood 95:309-313, 2000.

6. Di Raimondo F, Palumbo GA, Azzaro MP, et al: Angiogenesis inacute myeloid leukemia. Blood 96:3656-3657, 2000.

7. Di Raimondo F, Azzaro MP, Palumbo G, et al: Angiogenicfactors in multiple myeloma: Higher levels in bone marrow than in peripheralblood. Haematologica 85:800-805, 2000.

8. Di Raimondo F, Fichera E, Azzaro MP, et al: Low serum levelsof vascular endothelial growth factor (VEGF) and hepatocyte growth factor (HGF)in IFN-treated chronic myeloid leukemia (CML) patients (abstract 4825). Blood96:253b, 2000.

9. Aguayo A, Estey E, Kantarjian H, et al: Cellular vascularendothelial growth factor is a predictor of outcome in patients with acutemyeloid leukemia. Blood 94:3717-3721, 1999.

10. Pruneri G, Bertolini F, Soligo D, et al: Angiogenesis inmyelodysplastic syndromes. Br J Cancer 81:1398-1401, 1999.

11. Verstovsek S, Estey E, Manshouri T, et al: High expressionof the receptor tyrosine kinase Tie-1 in acute myeloid leukemia andmyelodysplastic syndrome. Leuk Lymphoma 42:511-516, 2001.

12. Lundberg LG, Lerner R, Sundelin P, et al: Bone marrow inpolycythemia vera, chronic myelocytic leukemia, and myelofibrosis has anincreased vascularity. Am J Pathol 157:15-19, 2000.

13. Mesa RA, Hanson CA, Rajkumar SV, et al: Evaluation andclinical correlations of bone marrow angiogenesis in myelofibrosis with myeloidmetaplasia. Blood 96:3374-3380, 2000.

14. Kini AR, Kay NE, Peterson LC: Increased bone marrowangiogenesis in B cell chronic lymphocytic leukemia. Leukemia 14:1414-1418,2000.

15. Molica S, Vitelli G, Levato D, et al: Increased serum levelsof vascular endothelial growth factor predict risk of progression in earlyB-cell chronic lymphocytic leukaemia. Br J Haematol 107:605-610, 1999.

16. Molica S, Santoro R, Digiesi G, et al: Vascular endothelialgrowth factor isoforms 121 and 165 are expressed on B-chronic lymphocyticleukemia cells. Haematologica 85:1106-1108, 2000.

17. Chen CA, Cheng WF, Lee CN, et al: Serum vascular endothelialgrowth factor in epithelial ovarian neoplasms: correlation with patientsurvival. Gynecol Oncol 74:235-240, 1999.

18. Duensing S, Atzpodien J: Increased intracellular and plasmalevels of basic fibroblast growth factor in B-cell chronic lymphocytic leukemia.Blood 85:1978-1980, 1995.

19. Ferrajoli A, Manshouri T, Estrov Z, et al: High levels ofvascular endothelial growth factor receptor-2 correlate with shortened survivalin chronic lymphocytic leukemia. Clin Cancer Res 7:795-799, 2001.

20. Gruber G, Schwarzmeier JD, Shehata M, et al: Basicfibroblast growth factor is expressed by CD19/CD11c-positive cells in hairy cellleukemia. Blood 94:1077-1085, 1999.

21. Rajkumar SV, Leong T, Roche PC, et al: Prognostic value ofbone marrow angiogenesis in multiple myeloma. Clin Cancer Res 6:3111-3116, 2000.

22. Ribatti D, Vacca A, Nico B, et al: Bone marrow angiogenesisand mast cell density increase simultaneously with progression of human multiplemyeloma. Br J Cancer 79:451-455, 1999.

23. Vacca A, Ribatti D, Presta M, et al: Bone marrowneovascularization, plasma cell angiogenic potential, and matrixmetalloproteinase-2 secretion parallel progression of human multiple myeloma.Blood 93:3064-3073, 1999.

24. Salven P, Teerenhovi L, Joensuu H: A high pretreatment serumbasic fibroblast growth factor concentration is an independent predictor of poorprognosis in non-Hodgkin’s lymphoma. Blood 94:3334-3339, 1999.

25. Salven P, Orpana A, Teerenhovi L, et al: Simultaneouselevation in the serum concentrations of the angiogenic growth factors VEGF andbFGF is an independent predictor of poor prognosis in non-Hodgkin lymphoma: Asingle-institution study of 200 patients. Blood 96:3712-3718, 2000.

26. Foss HD, Araujo I, Demel G, et al: Expression of vascularendothelial growth factor in lymphomas and Castleman’s disease. J Pathol183:44-50, 1997.

27. Rajkumar SV, Kyle RA: Angiogenesis in multiple myeloma.Semin Oncol 28:560-564, 2001.

28. Gille H, Kowalski J, Li B, et al: Analysis of biologicaleffects and signaling properties of Flt-1 (VEGFR-1) and KDR (VEGFR-2). Areassessment using novel receptor-specific vascular endothelial growth factormutants. J Biol Chem 276:3222-3230, 2001.

29. Kliche S, Waltenberger J: VEGF receptor signaling andendothelial function. IUBMB Life 52:61-66, 2001.

30. Yancopoulos GD, Davis S, Gale NW, et al: Vascular-specificgrowth factors and blood vessel formation. Nature 407:242-248, 2000.

31. Aguayo A, Kantarjian H, Manshouri T, et al: Angiogenesis inacute and chronic leukemias and myelodysplastic syndromes. Blood 96:2240-2245,2000.

32. Rodewald HR, Sato TN: Tie-1, a receptor tyrosine kinaseessential for vascular endothelial cell integrity, is not critical for thedevelopment of hematopoietic cells. Oncogene 12:397-404, 1996.

33. Fiedler W, Graeven U, Ergun S, et al: Vascular endothelialgrowth factor, a possible paracrine growth factor in human acute myeloidleukemia. Blood 89:1870-1875, 1997.

34. Aguayo A, Manshouri T, O’Brien S, et al: Clinicalrelevance of Flt-1 and Tie-1 angiogenesis receptors expression in B-cell chroniclymphocytic leukemia (CLL). Leuk Res 25:279-285, 2001.

35. Neben K, Moehler T, Egerer G, et al: High plasma basicfibroblast growth factor concentration is associated with response tothalidomide in progressive multiple myeloma. Clin Cancer Res 7:2675-2681, 2001.

36. Borset M, Seidel C, Hjorth-Hansen H, et al: The role ofhepatocyte growth factor and its receptor c-Met in multiple myeloma and otherblood malignancies. Leuk Lymphoma 32:249-256, 1999.

37. Chiarugi V, Magnelli L, Gallo O: COX-2, iNOS and p53 asplay-makers of tumor angiogenesis. Int J Mol Med 2:715-719, 1998.

38. Fosslien E: Biochemistry of cyclooxygenase (COX)-2inhibitors and molecular pathology of COX-2 in neoplasia. Crit Rev Clin Lab Sci37:431-502, 2000.

39. Fosslien E: Molecular pathology of cyclooxygenase-2 inneoplasia. Ann Clin Lab Sci 30:3-21, 2000.

40. Bertagnolli MM: Cyclooxygenase-2 as a target for preventionof colorectal cancer. Curr Oncol Rep 1:173-178, 1999.

41. Adams J: Proteasome inhibition in cancer: development ofPS-341. Semin Oncol 28:613-619, 2001.

42. Hideshima T, Richardson P, Chauhan D, et al: The proteasomeinhibitor PS-341 inhibits growth, induces apoptosis, and overcomes drugresistance in human multiple myeloma cells. Cancer Res 61:3071-3076, 2001.

43. Chen Y, Wiesmann C, Fuh G, et al: Selection and analysis ofan optimized anti-VEGF antibody: Crystal structure of an affinity-matured Fab incomplex with antigen. J Mol Biol 293:865-881, 1999.

44. Presta LG, Chen H, O’Connor SJ, et al: Humanization of ananti-vascular endothelial growth factor monoclonal antibody for the therapy ofsolid tumors and other disorders. Cancer Res 57:4593-4599, 1997.

45. Mendel DB, Schreck RE, West DC, et al: The angiogenesisinhibitor SU5416 has long-lasting effects on vascular endothelial growth factorreceptor phosphorylation and function. Clin Cancer Res 6:4848-4858, 2000.

46. Mendel DB, Laird AD, Smolich BD, et al: Development ofSU5416, a selective small molecule inhibitor of VEGF receptor tyrosine kinaseactivity, as an anti-angiogenesis agent. Anticancer Drug Des 15:29-41, 2000.

47. Taylor ML, Metcalfe DD: Kit signal transduction. HematolOncol Clin North Am 14:517-535, 2000.

48. Blair A, Sutherland HJ: Primitive acute myeloid leukemiacells with long-term proliferative ability in vitro and in vivo lack surfaceexpression of c-kit (CD117). Exp Hematol 28:660-671, 2000.

49. Smolich BD, Yuen HA, West KA, et al: The antiangiogenicprotein kinase inhibitors SU5416 and SU6668 inhibit the SCF receptor (c-kit) ina human myeloid leukemia cell line and in acute myeloid leukemia blasts. Blood97:1413-1421, 2001.

50. Mesters RM, Padro T, Bieker R, et al: Stable remission afteradministration of the receptor tyrosine kinase inhibitor SU5416 in a patientwith refractory acute myeloid leukemia. Blood 98:241-243, 2001.

51. Wood JM, Bold G, Buchdunger E, et al: PTK787/ZK 222584, anovel and potent inhibitor of vascular endothelial growth factor receptortyrosine kinases, impairs vascular endothelial growth factor-induced responsesand tumor growth after oral administration. Cancer Res 60:2178-2189, 2000.