Immune Dysfunction in Cancer Patients

January 1, 2002

Immune deficiency in cancer patients is well documented, and tumor cells have developed a variety of cellular and molecular mechanisms to avoid antitumor immune responses. These mechanisms include defective presentation of tumor antigens on the cell surface and/or an inability of the host to effectively recognize these cells and target them for destruction. Tumor-induced defects are known to occur in all major branches of the immune system. The continuous administration of vascular endothelial growth factor (VEGF), a factor produced by most solid tumors, inhibits the functional maturation of dendritic cells, significantly decreases T-cell to B-cell ratios in the peripheral lymphoid organs, and induces rapid and dramatic thymic atrophy in tumor-bearing animals. VEGF is abundantly expressed by a large percentage of solid tumors, and defective antigen presentation, T-cell defects, and premature thymic atrophy are known to occur in cancer patients and tumor-bearing animals. This review will encompass the major mechanisms responsible for tumor evasion of immune surveillance and highlight a role for VEGF as a principal contributor to tumor-associated immune deficiencies. [ONCOLOGY 16(Suppl 1):11-18, 2002]

ABSTRACT: Immune deficiency in cancer patients is well documented, and tumor cells have developed a variety of cellular and molecular mechanisms to avoid antitumor immune responses. These mechanisms include defective presentation of tumor antigens on the cell surface and/or an inability of the host to effectively recognize these cells and target them for destruction. Tumor-induced defects are known to occur in all major branches of the immune system. The continuous administration of vascular endothelial growth factor (VEGF), a factor produced by most solid tumors, inhibits the functional maturation of dendritic cells, significantly decreases T-cell to B-cell ratios in the peripheral lymphoid organs, and induces rapid and dramatic thymic atrophy in tumor-bearing animals. VEGF is abundantly expressed by a large percentage of solid tumors, and defective antigen presentation, T-cell defects, and premature thymic atrophy are known to occur in cancer patients and tumor-bearing animals. This review will encompass the major mechanisms responsible for tumor evasion of immune surveillance and highlight a role for VEGF as a principal contributor to tumor-associated immune deficiencies. [ONCOLOGY 16(Suppl 1):11-18, 2002]

An essential function of the immune system is theability to defend against pathogenic infections. Immune cells can identifyforeign antigens expressed on the surface of an infected cell, such as viral orbacterial proteins, and target these cells for destruction. Mutations and/oralterations in normal cellular proteins that arise in a cancerous cell alsoresult in the display of unique antigens on the surface of these cells. Whenfully functional, the immune system has the capability to identify cancer cellsas "non-self" and eliminate them from the body. It is self-evident,however, that clinically apparent tumors avoid effective antitumor immuneresponses; in fact, cancer patients often exhibit an immune-compromisedphenotype that extends beyond an inability to recognize tumor antigens.[1]

Tumor cells have developed a variety of cellular and molecular mechanisms toavoid antitumor immune responses,[2-8] including host alterations in T-cellreceptor/CD3 complex expression and function, decreased major and minorhistocompatibility complex expression by the tumor, and loss of tumor epitopes.Virtually all branches of the immune system can be affected. Tumor cells alsosecrete a variety of soluble factors that are capable of inhibiting immune cellfunction, such as interleukin (IL)-10, tumor necrosis factor (TNF), transforminggrowth factor-beta (TGF-beta), and vascular endothelial growth factor (VEGF).The effects of these factors appear to be twofold: to inhibit immune celleffector function and to impair the development of immune cells by acting onearlier stages of immunopoiesis.

VEGF and its receptors have profound effects on the early development anddifferentiation of both vascular endothelial and hematopoietic progenitors.[9]It induces proliferation of mature endothelial cells and is an importantcomponent in the formation of tumor neovasculature.[10] VEGF is abundantlyexpressed by a large percentage of solid tumors; this overexpression is closelyassociated with a poor prognosis.[11,12] Some of the earliest hematopoieticprogenitors express receptors for VEGF[13]; we have demonstrated that VEGFcauses a defect in the functional maturation of dendritic cells fromprogenitors, resulting in defective antigen presentation. This developmentaldefect is associated with impaired activation of NF-kappaB.[14-17]

In addition to defects in the myeloid lineage, VEGF also plays a key role inmediating the development of lymphoid lineage cells. VEGF induces dramaticthymic atrophy resulting in decreased numbers of mature T cells in theperiphery, and the loss of the effector cells may also significantly impair anantitumor response (unpublished data).

This article will attempt to provide the reader with an understanding of themajor problems that can lead to a failure of antitumor immune induction, withspecial emphasis on our ongoing research into the important role VEGF plays inmediating this effect. We demonstrate that VEGF is not only important for tumorvascularization, but is also a key factor produced by solid tumors to inhibitrecognition and destruction of tumor cells by the immune system.

Defective Antigen Presentation

A primary role of the immune system is to distinguish "self" from"non-self" proteins. Foreign antigens expressed by viruses or bacteriacan be presented on the surface of an infected cell, and identify that cell asnon-self for destruction by the immune system. Similarly, unique or alteredversions of normal cellular proteins produced by tumor cells can be presented tocytotoxic T cells, resulting in a host response against the tumor. Chemical orphysical carcinogens can induce tumor antigens[18] or they may originate inspontaneous tumors. To date, a large number of tumor antigens have beenidentified.[18-23] These endogenous tumor antigens may be derived from fetal orembryonic genes, mutant oncogenes, or oncogenic viral genes such as humanpapillomavirus.

The display of tumor antigens on the cell surface is essential for therecognition and destruction of a tumor cell by the immune system. Tumor orforeign antigens must be degraded, along with normal cellular proteins, intosmall peptides by the proteosome. These peptides associate with class I MHC (MHC-I)in the lumen of the endoplasmic reticulum and are transported to the cellsurface for presentation to CD8-positive cytotoxic T cells. In cases where astructural defect has occurred within the tumor cell, a genetic mutation isoften responsible for disrupting the normal display of tumor antigens on thecell surface. These mutations may result in the inability of a cell to producetransporter molecules, such as TAP1, or other molecules essential for thisprocess, such as MHC-I or beta-2-microglobin, and will lead to a failure of thecell to present all antigens. However, structural defects of this nature areonly found in approximately 5% to 10% of human tumors, and the majority of humantumors are ineffective at directly inducing an immune response despite adequatedisplay of tumor antigens on their cell surface.

What causes this lack of an antitumor immune response in the remaining 90% to95% of human tumors? Induction of an effective immune response is a complexprocess that involves many cell types and cytokine mediators. Tumor-bearinghosts have acquired deficiencies in several of the host elements responsible forthis induction. We have found that defects in both myeloid lineage and lymphoidlineage cells are major components of this problem, and the remainder of thisarticle will focus on our studies in this area.

Dendritic Cells in Antitumor Immune Responses

Professional antigen-presenting cells are responsible for the presentation oftumor antigens to both B and T lymphocytes, and can therefore induce bothhumoral and cell-mediated responses against a tumor (Figure1). Several studieshave described the defects in the function of antigen-presenting cells intumor-bearing hosts.[24-26] Dendritic cells are the most potentantigen-presenting cells; for this reason, they are potential targets for tumorvaccines and immunotherapies. Because of the central role that dendritic cellsplay in induction of antitumor immunity, research in our laboratory has focusedon the hypothesis that defects in dendritic cell function may potentiallyaccount for the immunoresistance of certain tumors.

Tumor-derived factors with the potential to interfere with the development orfunction of immune cells play an important role in the escape of tumors fromnormal immune surveillance. We have demonstrated that tumor cells secretesoluble factors that can inhibit the maturation of CD34-positive hematopoieticprogenitor cells into functional dendritic cells when cultured in vitro.[14,27]CD34-positive hematopoietic progenitor cells were isolated from human cord bloodand cultured in vitro in the presence of granulocyte-macrophagecolony-stimulating factor (GM-CSF), IL-4, and TNF-alpha.

Tumor-cell supernatants, derived from colon and breast adenocarcinoma celllines, were added to hematopoietic progenitor cells to determine the effect oftumor-derived soluble factors on dendritic cell maturation in vitro. Dendriticcell function was then measured by two distinct assays: (1) the ability ofmature dendritic cells to stimulate proliferation of allogeneic T cells in mixedleukocyte reactions; and (2) the ability to take up fluorescein isothiocyanate (FITC)-dextran.Using both assays, we found that tumor-cell supernatants dramatically reduceddendritic cell function. Dendritic cells obtained after the culture ofhematopoietic progenitor cells with tumor-cell supernatants were not onlyfunctionally impaired, but also morphologically distinct from mature dendriticcells.

Overall, the number of mature dendritic cells present in the tumor-cellsupernatant cultures were reduced two- to threefold. These cells expressedreduced levels of mature dendritic cell surface markers and exhibited severalcharacteristics of immature myeloid cells. Tumor-cell supernatants did notinhibit proliferation of CD34-positive progenitors, nor did they affect thetotal number of CD34-positive or CD34-positive/CD38-negative progenitor cells,indicating that tumor-cell supernatant-induced defects did not result from theloss of multipotent progenitor cells. Furthermore, inhibition of dendritic cellfunction was observed only when tumor-cell supernatants were added within thefirst 4 days of in vitro culture, indicating an effect on early dendritic celldevelopment.[14]

Size fractionation experiments demonstrated that dendritic cell-inhibitoryaction was restricted to the 30 to 50 kD size fraction of tumor-cellsupernatants. Neutralizing antibodies to proteins within this size range, andknown to be produced by tumor cells, were added to mixed leukocyte reactions inan attempt to identify the dendritic cell-inhibitory factor. Neutralizingantibodies to VEGF, but not antibodies against TGF-beta, IL-10, or c-kit,blocked the ability of dendritic cells to stimulate proliferation of allogeneicT cells[14] (Figure 2). Furthermore, there was a tight correlation between VEGFconcentrations and the inhibitory activity of tumor-cell supernatants in 12tumor cell lines observed. These data indicate that inhibition of dendritic cellfunction by tumor-cell supernatants is substantially mediated by VEGF.

The Role of VEGF

The VEGF protein (34 to 43 kD) binds specifically to endothelial cells andstimulates angiogenesis and endothelial cell migration in vivo.[28] VEGF isproduced by a large majority of solid tumors[28] and is found at elevated levelsin the serum of cancer patients.[29] This abnormal VEGF expression plays animportant role in the formation of tumor neovasculature, contributing to thegrowth of tumors.[10] Consistent with these findings, VEGF production by tumorsis closely associated with a poor prognosis.[12,30]

VEGF binds to two primary receptors—VEGFR-1 (Flt-1) and VEGFR-2 (Flk-1/KDR).VEGF and its receptors have profound effects on the early development anddifferentiation of both vascular and hematopoietic progenitors.[9] VEGFR-1expression by CD34-positive hematopoietic progenitor cells has been confirmed byreverse transcription polymerase chain reaction analysis.[14,31] VEGFR-2 is alsoexpressed by hematopoietic progenitor cells,[32] and can be used as a marker forhematopoietic stem cells derived from hemangioblast progenitors.[13] Thehemangioblast is an as-yet undiscovered progenitor cell population that isthought to give rise to both the endothelial and hematopoietic lineages.

A functional role for VEGF has been shown in hematopoietic cells as well. Theessential role of VEGF in hematopoietic differentiation is underscored by thefact that heterozygous knockouts of VEGF are embryonic lethal and have defectiveblood island formation.[9] VEGF has also been shown to suppressradiation-induced apoptosis in normal hematopoietic stem cells,[32] suggestingthat VEGF may act as a survival factor during hematopoiesis. In addition, VEGFmay promote differentiation of certain hematopoietic lineages. Broxmeyer et alhave demonstrated that VEGF enhances colony formation by mature subsets ofgranulocyte-macrophage progenitor cells while inhibiting formation of coloniesnot stimulated by these factors.[33] Based on these past studies, wehypothesized that VEGF provides dual differentiation signals duringhematopoiesis—promoting the survival of certain hematopoietic lineages butinhibiting the development and maturation of others.

VEGF Inhibits Development of Dendritic Cells

The above studies suggested that VEGF is a tumor-derived factor at leastpartly responsible for the defective maturation of dendritic cells in vitro.This finding may provide a mechanism to explain tumor evasion of immunesurveillance. In order to confirm the role of VEGF on dendritic cell maturationand function in vivo, mice were given a continuous infusion of VEGF viasubcutaneous osmotic pumps (50 ng/h for a period of 28 days). This techniqueproduces steady-state serum concentrations of VEGF between 120 and 160 pg/mL,well within the pathophysiological range observed in advanced-stage cancerpatients.[34]

Continuous infusion of VEGF into mice resulted in a substantial inhibition ofdendritic cell development.[15] The ability of splenic dendritic cells topresent antigen to allogeneic T cells was dramatically reduced following 14 daysof VEGF infusion. Stimulation of a primary immune response to influenza virus bydendritic cells and the ability to ingest FITC-dextran were also dramaticallyimpaired. There were also decreased numbers of Langerhans’ cells—specializeddendritic cells in the skin—which were impaired in their ability to take upantigen and migrate to lymph nodes.[15,16]

While defects in dendritic cell function were observed following 14 days VEGFinfusion, we only observed significant differences in hematopoietic cellpopulations following 28 days of VEGF infusion. Fluorescence-activated cellsorting analysis revealed a decreased percentage of CD11c-positive/B7-2-positivedendritic cells in both the spleen and lymph nodes following 28 days of VEGFinfusion. There was a twofold reduction in mature dendritic cells purified fromlymph nodes and a more than fourfold reduction in dendritic cells isolated fromthe spleen. The proportion of lymphocytes in the spleen was also reduced, anddespite a fourfold increase in overall splenic cellularity, the number of Bcells and T cells was reduced. At the same time, the number of Gr1-positivemyeloid cells and Ter119-positive erythroid cells was markedly increased.[15]

The fraction of lymphocytes present in the lymph nodes of VEGF-infused mice(28 days) was unaffected, but there was a dramatic shift in B-cell to T-cellratio, with a greater than threefold increase in the percentage of B cells.Immature myeloid cells and erythroid cells were also increased in the lymphnodes. There was no apparent change in percentage of stem cells or endothelialcells in these tissues.[15] Overall, our data indicate that continuous infusionof VEGF results in the inhibition of dendritic cell function, alterations inlymphocyte numbers, and the accumulation of immature myeloid cells andgranulocytes. Similar to what we have observed in vitro, the effects of VEGF ondendritic cell development are evident within the first few days of infusion asmonitored by colony-forming assays on Lin-negative hematopoietic progenitorcells.[15]

Dendritic Cell Defects

Defective dendritic cell function has been observed previously in both cancerpatients and tumor-bearing mice.[24-26] However, the mechanisms that lead todendritic cell dysfunction, as well as the clinical significance behind thisphenomenon, remain unclear. In a study of 93 patients with breast, head/neck,and lung cancer, the clinical correlation between dendritic cell dysfunction anddisease progression was observed.[34] Specifically, the number of maturedendritic cells was drastically reduced in the peripheral blood andtumor-draining lymph nodes of cancer patients. These results indicate a systemicimmune deficiency in the cancer patients rather than a localized tumor effect.

The dendritic cell decreases were also associated with an increase inimmature myeloid cells. The presence of these immature hematopoietic cellscorrelated closely to the stage and duration of the disease, and defects werepartially corrected with successful surgical resection of the tumor. Thepresence of immature hematopoietic cells in the peripheral blood of cancerpatients was also associated with increased plasma levels of VEGF, but not IL-6,GM-CSF, macrophage colony-stimulating factor (M-CSF), IL-10, or TGF-beta.Together these findings suggest that VEGF is one of the primary soluble factorsproduced by tumors to suppress immune function.

Antibodies to VEGF Enhance Immunotherapy

Inadequate function of dendritic cells in tumor-bearing hosts may be onefactor that compromises the efficacy of cancer immunotherapy. If VEGF is theprimary factor responsible for impaired maturation and function of dendriticcells, then inhibition of VEGF should correct this defect in cancer patients. Inaddition to their antiangiogenic activities, therapeutic strategies targetingVEGF may lead to improved immune function. We have investigated a novelcombination of VEGF-targeted antiangiogenic therapy and immunotherapy using twosubcutaneous tumor models in mice: D459 cells, expressing mutant human p53, andMethA sarcoma, with point mutations in the endogenous p53 gene.[35]

Therapy with antimouse VEGF antibody (10 mg intraperitoneally twice a weekover 4 weeks) was initiated when tumors became palpable. Anti-VEGF therapysignificantly improved the number and function of lymph node- andspleen-derived dendritic cells isolated from these tumor-bearing animals, thoughtreatment of tumors with anti-VEGF antibody alone did not affect the rate oftumor growth. Anti-VEGF antibody therapy also improved the efficacy of cancerimmunotherapy when given in conjunction with peptide-pulsed dendritic cellscorresponding to the mutation-specific p53 peptides.[35]

Limited data also indicate that antibodies to VEGF are able to substantiallycorrect dendritic cell defects in cancer patients. In a small pilot study, threepatients with metastatic lung cancer were given antibodies to VEGF. A completereversal of all dendritic cell maturation defects was observed in all threepatients.[34] Further clinical trials investigating the use of VEGF-targetedtherapeutics are ongoing and utilize a variety of different strategies forinterruption of VEGF signaling. These strategies include monoclonal antibodies,ribozymes, and specific receptor tyrosine-kinase inhibitors generally incombination with chemotherapy and targeting various disease sites. Morepertinent to this discussion, combining anti-VEGF therapies with immunotherapiesmay dramatically improve their efficacy; several such trials are planned.

T Cells in Antitumor Immune Responses

In addition to defects in dendritic cells and other myeloid cells,tumor-associated immune deficiencies have also been observed in the thymus and Tcells.[2-8] While thymic atrophy accompanies normal aging,[36] a high incidenceof premature thymic involution is also seen in patients with childhoodmalignancies[37]; this often rebounds after curative treatment. This can bemodeled in mice transplanted with mammary adenocarcinomas, which demonstraterapid thymic involution associated with depletion and/or alterations ofthymocyte subpopulations.[38-41] The mechanism of this cancer-associated thymicatrophy, and more generally the factors responsible for lineage commitment,migration to the thymus, and progression through thymocyte developmentalcheckpoints, remains poorly understood. In addition to dendritic cell defects,non-tumor-bearing mice treated with a continuous infusion of recombinant VEGFhave a decreased number of T cells in their lymph nodes and spleen,[15] and thusVEGF represents a candidate mediator of the tumor-associated thymic defect.

After 3 to 4 weeks of VEGF treatment as described above, we have observed adramatic reduction in the size of the thymus and a striking decrease inthymocyte cellularity as compared to age-matched, phosphate-buffered saline(PBS)-infused controls (unpublished data). The normally clear distinctionbetween cortical and medullary regions of the thymus is no longer present, andbroad hypocellular areas containing collagen, fibroblasts, and dilated bloodvessels separated the remaining follicles. The relative percentage of boththymic epithelial cells and vascular structures is increased in VEGF-treatedlobes, indicating that cell loss is primarily restricted to the lymphocytepopulations. There is no significant difference in thymus size, structure, orcellularity following infusions of less than 14 days or with PBS infusions,suggesting that the observed effects were not due to postsurgical stress.

Thymic atrophy accompanies a dramatic drop in total thymocytes, and this lossis disproportionately seen in the CD4-positive/CD8-positive thymocytepopulations. It does not appear to be associated with a significant increase inapoptosis or cell cycle arrest, and VEGF has no effect on thymocyte developmentin fetal thymic organ culture (unpublished data). All of our data indicate thatVEGF acts on thymic progenitors rather than directly on the thymus itself. Wepropose that treatment with VEGF results in defective seeding of the thymus bybone marrow-derived progenitors, and as these earliest thymocytes fail toreplace maturing T cells migrating to the periphery, a depletion of totalthymocytes results.

We also demonstrate that hematopoietic progenitor cells from the bone marrowof VEGF-treated mice, when transferred into an irradiated host, reconstitute thethymus 2.5- to 3-fold more efficiently than control hematopoietic progenitorcells. This indicates an increased number of functional thymus-directedprogenitors in the marrow of VEGF-treated animals. Thus VEGF appears to induce ablock in the differentiation of thymic precursors from lymphoid progenitors inthe marrow, resulting in the accumulation of cells upstream from this block.These data are consistent with our dendritic cell data, and strengthen thehypothesis that VEGF affects the development and differentiation of multiplehematopoietic lineages.

VEGF Signaling in Hematopoietic Progenitor Cells

We have demonstrated that VEGF affects the ability of hematopoieticprogenitor cells to differentiate into functional dendritic cells during theearly stages of hematopoiesis, and induces dramatic thymic atrophy and defectiveT-cell maturation in vivo. Significant research efforts have been focused on themolecular mechanisms and signaling pathways that are responsible for thiseffect. We have shown that VEGF binds to specific receptors on the surface ofhematopoietic progenitor cells.[17] This binding can be successfully competed byplacental growth factor, indicating that signaling is at least in partaccomplished by binding of VEGF to the VEGFR-1 (Flt-1) receptor. Additionally,the number of binding sites available for VEGF decreased with dendritic cellmaturation and correlated with decreased levels of VEGFR-1 mRNA expression inthe late-stage cells.

The NF-kappaB signaling pathway has been implicated in the differentiationand function of a variety of immune cell lineages.[42-47] There are five knownNF-kappaB/Rel proteins: Rel-A (p65), Rel-B, c-Rel, p50, and p52. Hetero- andhomodimers of members of the NF-kappaB family of transcription factors aresequestered in the cytoplasm of cells by inhibitor proteins: I-kappaB-alpha, I-kappaB-beta,I-kappaB-epsilon, I-kappaB-gamma, and Bcl-3. Activation of NF-kappaB is mediatedby phosphorylation, ubiquitination, and degradation of I-kappaB, followed bytranslocation of NF-kappaB to the nucleus. Known signaling pathwaysdemonstrating the importance of NF-kappaB signaling in the immune system includeTNF-alpha, IL-1alpha, antigen receptors on B and T lymphocytes, andcostimulatory molecules such as CD28 and CD40.[45,47-50]

As early as 30 minutes after TNF-alpha induction,[17] VEGF inhibits TNF-alpha-inducedbinding of NF-kappaB to target DNA sequences in hematopoietic progenitor cells(see Figure 3).[15] VEGF also significantly inhibits NF-kappaB-dependentactivation of reporter gene transcription in dendritic cells during the first 24hours in culture. In addition, VEGF treatment of hematopoietic progenitor cellsreduced levels of Rel-Band c-Rel mRNA within 7 to 10 days of culture. To assess the biologicsignificance of VEGF inhibition of NF-kappaB signaling, we employed anadenovirus that encodes a dominant I-kappaB inhibitor. This inhibitor, termed I-kappaB-DN,effectively blocks NF-kappaB activity in hematopoietic progenitor cells at theirearly stages of differentiation.

Importantly, expression of I-kappaB-DN reproduces the inhibitory effects ofVEGF on hematopoietic progenitor cell differentiation. Substitution of VEGF withplacental growth factor also resulted in a reduction of NF-kappaB nuclearlocalization and binding to target sequences.[17] From these data, we concludethat VEGF blocks NF-kappaB activation by TNF-alpha, and NF-kappaB plays asignificant role in the maturation of dendritic cells from hematopoieticprogenitor cells. These findings also suggest a mechanism by which tumor-derivedsoluble factors may directly down-regulate immune responses to tumor antigens.

More recent studies in our laboratory have shown that TNF-alpha treatment ofdendritic cells induces the nuclear translocation of NF-kappaB complexescontaining Rel-A (unpublished results). In contrast, VEGF is able to signalindependently through the NF-kappaB signaling pathway, but induction of NF-kappaBsignaling is mediated by distinct Rel subunits. VEGF treatment alone induces thetranslocation of Rel-B-containing complexes. Surprisingly, Rel-B induction wasnot necessary for VEGF-mediated inhibition of NF-kappaB signaling by TNF-alpha,as similar effects were observed in dendritic cells from Rel-B-/- knockout micecompared with wild type and heterozygous (Rel-B+/-) controls.

Nuclear translocation of NF-kappaB is induced by degradation of associated I-kappaBvia activation of IKK (inhibitor of -kappaB kinase). Incubation with VEGFdecreased TNF-alpha-induced IKK activation, decreased phosphorylation of I-kappaB,and impaired degradation of I-kappaB-alpha and I-kappaB-epsilon in culturedhematopoietic progenitor cells. In the absence of TNF-alpha, VEGF had no effecton I-kappaB phosphorylation or degradation. At present, the mechanisms by whichVEGF inhibits TNF-alpha-induced IKK activation remain unknown (unpublishedresults).

Both VEGFR-1 and VEGFR-2 are members of the fms family of tyrosine kinasereceptors. As such, it was surprising that SU5416, a potent inhibitor of VEGFR-1(Flt-1) and VEGFR-2 (Flk-1/KDR) tyrosine kinase activity, failed to reverse theinhibitory effect of VEGF on NF-kappaB activation. These findings indicate thatthe effects of VEGF are mediated by a non-tyrosine kinase signaling pathway.The argument for a non-tyrosine kinase pathway for VEGF signaling is furtherstrengthened by knockout studies of VEGFR-1 and VEGFR-2. Null mutations forthese receptors are both embryonic lethal,[10,51] whereas mice harboring adeletion in the intracellular domain of VEGFR-1 are viable.[52] The lattermutant strain completely lacks the VEGFR-1 tyrosine kinase domain, but does notdisplay any appreciable defects in development.

In addition, angiogenesis and vascularization during adulthood appearsunaffected in VEGFR-1 intracellular domain knockouts, indicating that VEGFR-1has an essential function in development that is unrelated to its tyrosinekinase activity. Further investigation is necessary to determine whether micelacking the cytoplasmic domain of VEGFR-1 exhibit defects in hematopoiesis; thisinformation is important in considering the use of anti-VEGFR-1-targetedtherapies. The presence of an essential VEGF signaling pathway that isindependent of VEGFR-1 tyrosine kinase activity may provide a mechanisticexplanation for the dual activities of VEGF. One pathway may be responsible forthe angiogenic properties of VEGF, while an alternate signaling cascade mightmediate the suppression of immune responses to tumors. A dissection of VEGFsignaling is an important goal for the development of anti-VEGF adjuvanttherapies.

Summary

We have shown that VEGF promotes tumor formation by several modes of action.In addition to its well-documented role in tumor neo-vascularization, VEGF alsomediates tumor evasion of normal immune surveillance by inhibiting thedevelopment of both dendritic cells and T cells. Other hematopoietic lineagesare conversely effected. These defects manifest as an impairment of the immunesystem to recognize tumor antigens presented on the cell surface, and thereforeallow the tumor to avoid destruction by immune effector cells. Suppression ofdendritic cell development by VEGF is observed both in vitro and in vivo andoccurs early on in the differentiation of dendritic cells from hematopoieticprogenitors. Defects in T-cell development are primarily manifested in adramatic induction of thymic atrophy and loss of thymocyte cellularity.

Development of both lineages appears to be inhibited by activity of VEGF onearly hematopoietic progenitor cells in the bone marrow. The importance of VEGFas a tumor-derived soluble factor that is responsible for inhibition of immunecell function is underscored in both the clinical and preclinical experimentalmodels. Antibodies to VEGF rescue dendritic cell maturation defects in cancerpatients and tumor-bearing animals, and thymic atrophy associated withmalignancy is observed in both systems.

Dendritic cell effects are likely mediated by VEGF signaling through VEGFR-1,which is transduced via a non-tyrosine kinase-dependent pathway, and wehypothesize a similar signaling mechanism in lymphoid progenitors. Thefunctional outcome of VEGFR-1 signaling is impaired induction of NF-kappaB, atranscription factor family known to be important for hematopoietic development.Further investigation into the role of VEGF in immune evasion, its mechanisms ofaction, and clinical significance may provide new insights into the developmentof novel therapeutic strategies to bolster antitumor immunotherapy.

References:

1. Kavanaugh DY, Carbone DP: Immunologic dysfunction in cancer. Hem Onc ClinNorth Am 10:927-951, 1996.

2. Johnsen AK, Templeton, DJ, Sy M, et al: Deficiency of transporter forantigen presentation (TAP) in tumor cells allows evasion of immune surveillanceand increases tumorigenesis. J Immunol 163:4224-4231, 1999.

3. Finke J, Ferrone S, Frey A, et al: Where have all the T cells gone?Mechanisms of immune evasion by tumors. Immunol Today 20:158-160, 1999.

4. Antonia SJ, Extermann M, Flavell RA: Immunologic nonresponsiveness totumors. Crit Rev Oncog 9:35-41, 1998.

5. Kiessling R, Wasserman K, Horiguchi S, et al: Tumor-induced immunedysfunction [see comments]. Cancer Immunol Immunother 48:353-362, 1999.

6. Shu S, Plautz GE, Krauss JC, et al: Tumor immunology. JAMA 278:1972-1981,1997.

7. Pawelec G, Zeuthen J, Kiessling R: Escape from host-antitumor immunity.Crit Rev Oncog 8:111-141, 1997.

8. Markiewicz MA, Gajewski TF: The immune system as anti-tumor sentinel:molecular requirements for an anti-tumor immune response. Crit Rev Oncog10:247-260, 1999.

9. Ferrara N, Carver-Moore K, Chen H, et al: Heterozygous embryonic lethalityinduced by targeted inactivation of the VEGF gene. Nature 380:439-442, 1996.

10. Ferrara N, Davis-Smyth T: The biology of vascular endothelial growthfactor. Endocr Rev 18:4-25, 1997.

11. Toi M, Hoshina S, Takayanagi T, et al: Association of vascularendothelial growth factor expression with tumor angiogenesis and with earlyrelapse in primary breast cancer. Jpn J Cancer Res 85:1045-1049, 1994.

12. Toi M, Taniguchi T, Yamamoto Y, et al: Clinical significance of thedetermination of angiogenic factors. Eur J Cancer 32A:2513-2519, 1996.

13. Ziegler BL, Valtieri M, Porada GA, et al: KDR receptor: A key markerdefining hematopoietic stem cells. Science 285:1553-1558, 1999.

14. Gabrilovich DI, Chen HL, Girgis KR, et al: Production of vascularendothelial growth factor by human tumors inhibits the functional maturation ofdendritic cells. Nat Med 2:1096-1103, 1996.

15. Gabrilovich D, Ishida T, Oyama T, et al: Vascular endothelial growthfactor inhibits the development of dendritic cells and dramatically affects thedifferentiation of multiple hematopoietic lineages in vivo. Blood 92:4150-4166,1998.

16. Ishida T, Oyama T, Carbone DP, et al: Defective function of Langerhanscells in tumor-bearing animals is the result of defective maturation fromhemopoietic progenitors. J Immunol 161:4842-4851, 1998.

17. Oyama T, Ran S, Ishida T, et al: Vascular endothelial growth factoraffects dendritic cell maturation through the inhibition of nuclear factor-kappaB activation in hemopoietic progenitor cells. J Immunol 160:1224-1232, 1998.

18. Boon T: Toward a genetic analysis of tumor rejection antigens. Adv CancerRes 58:177-210, 1992.

19. Boon T, van der Bruggen P: Human tumor antigens recognized by Tlymphocytes. J Exp Med 183:725-729, 1996.

20. Boon T, Cerottini J-C, Van der Eynde B, et al: Tumor antigens recognizedby T lymphocytes. Annu Rev Immunol 12:337-365, 1994.

21. Ciernik IF, Carbone DP: Tumor suppressor gene-derived peptide antigens.Meth Enzymol 8:225-233, 1995.

22. Coulie PG, Lehmann F, Lethe B, et al: A mutated intron sequence codes foran antigenic peptide recognized by cytolytic T lymphocytes on a human melanoma.Proc Natl Acad Sci U S A 92:7976-7980, 1995.

23. Van Pel A, DePlaen E, Lurquin C, et al: Identification of genes encodingT cell defined tumor antigens. Int Symp Princess Takamatsu Cancer Res Fund19:255-263, 1988.

24. Alcalay J, Kripke ML: Antigen-presenting activity of draining lymph nodecells from mice painted with a contact allergen during ultravioletcarcinogenesis. J Immunol 146:1717-1721, 1991.

25. Erroi A, Sironi M, Chiaffarino F, et al: IL1 and IL6 released bytumor-associated macrophages from human ovarian carcinoma. Int J Cancer44:795-801, 1989.

26. Watson GA, Lopez DM: Aberrant antigen presentation by macrophages fromtumor-bearing mice is involved in the down-regulation of their T cell responses.J Immunol 155:3124-3134, 1995.

27. Gabrilovich DI, Nadaf S, Corak J, et al: Dendritic cells in anti-tumorimmune responses. II. Dendritic cells grown from bone marrow precursors, but notmature DC from tumor-bearing mice are effective antigen carriers in the therapyof established tumors. Cell Immunol 170:111-119, 1996.

28. Ferrara N, Houck K, Jakeman L, et al: Molecular and biological propertiesof the vascular endothelial growth factor family of proteins. Endocr Rev13:18-32, 1992.

29. Kondo S, Asano M, Matsuo K, et al: Vascular endothelial growthfactor/vascular permeability factor is detectable in the sera of tumor-bearingmice and cancer patients. Biochem Biophys Acta 1221:211-214, 1994.

30. Ellis LM, Fidler IJ: Angiogenesis and metastasis. Eur J Cancer 32A,2451-2460, 1996.

31. Hoehn GT, Stokland T, Amin S, et al: Tnk1: A novel intracellular tyrosinekinase gene isolated from human umbilical cord blood CD34+/Lin-/CD38-stem/progenitor cells. Oncogene 12:903-913, 1996.

32. Katoh O, Tauchi H, Kawaishi K, et al: Expression of the vascularendothelial growth factor (VEGF) receptor gene, KDR, in hematopoietic cells andinhibitory effect of VEGF on apoptotic cell death caused by ionizing radiation.Cancer Res 55:5687-5692, 1995.

33. Broxmeyer HE, Cooper S, Li ZH, et al: Myeloid progenitor cells regulatoryeffects of vascular endothelial cell growth factor. Int J Hematol 62:203-215,1995.

34. Almand B, Resser JR, Lindman B, et al: Clinical significance of defectivedendritic cell differentiation in cancer. Clin Cancer Res 6:1755-1766, 2000.

35. Gabrilovich DI, Ishida T, Nadaf S, et al: Antibodies to vascularendothelial growth factor enhance the efficacy of cancer immunotherapy byimproving endogenous dendritic cell function. Clin Cancer Res 5:2963-2970, 1999.

36. Pawelec G, Effros RB, Caruso C, et al: T cells and aging (update February1999). Front Biosci 4:D216-D269, 1999.

37. Zhang M: [The relationships between thymus, other immune organs andvarious diseases in children (analysis of 621 cases)]. Chung Hua Ping Li HsuehTsa Chih 18:92-95, 1989.

38. Adkins B, Charyulu V, Sun QL, et al: Early block in maturation isassociated with thymic involution in mammary tumor-bearing mice. J Immunol164:5635-5640, 2000.

39. Fu Y, Paul RD, Wang Y, et al: Thymic involution and thymocyte phenotypicalterations induced by murine mammary adenocarcinomas. J Immunol 143:4300-4307,1989.

40. Lee MY, Rosse C: Depletion of lymphocyte subpopulations in primary andsecondary lymphoid organs of mice by a transplanted granulocytosis-inducingmammary carcinoma. Cancer Res 42:1255-1260, 1982.

41. Thomas E, Smith DC, Lee MY, et al: Induction of granulocytic hyperplasia,thymic atrophy, and hypercalcemia by a selected subpopulation of a murinemammary adenocarcinoma. Cancer Res 45:5840-5844, 1985.

42. Grossmann M, Nakamura Y, Grumont R, et al: New insights into the roles ofReL/NF-kappa B transcription factors in immune function, hemopoiesis and humandisease. Int J Biochem Cell Biol 31:1209-1219, 1999.

43. Ghosh S, May MJ, Kopp EB: NF-kappa B and Rel proteins: evolutionarilyconserved mediators of immune responses. Annu Rev Immunol 16:225-260, 1998.

44. Wulczyn FG, Krappmann D, Scheidereit C: The NF-kappa B/Rel and I kappa Bgene families: mediators of immune response and inflammation. J Mol Med74:749-769, 1996.

45. Baeuerle PA, Henkel T: Function and activation of NF-kappa B in theimmune system. Annu Rev Immunol 12:141-179, 1994.

46. Stankovski I, Baltimore D: NF-kB activation: The IkB kinase revealed.Cell 91:299-302, 1997.

47. Thanos D, Maniatis T: NF-kB: A lesson in family values. Cell 80:529-532,1995.

48. Verma IM, Stevenson JK, Schwarz EM, et al: Rel/NF-kappa B/I kappa Bfamily: Intimate tales of association and dissociation. Genes Dev 9:2723-2735,1995.

49. Baeuerle PA, Baltimore D: NF-kappa B: Ten years after. Cell 87:13-20,1996.

50. Baldwin AS: The NF-kappa B and I kappa B proteins: New discoveries andinsights. Annu Rev Immunol 14:649-683, 1996.

51. Neufeld G, Cohen T, Gengrinovitch S, et al: Vascular endothelial growthfactor (VEGF) and its receptors. FASEB J 13:9-22, 1999.

52. Hiratsuka S, Minowa O, Kuno J, et al: Flt-1 lacking the tyrosine kinasedomain is sufficient for normal development and angiogenesis in mice. Proc NatlAcad Sci U S A 95:9349-9354, 1998.