Over the last 2 decades, therapies that target tumor and endothelial cell surface receptors or their ligands have emerged as valuable treatment options in clinical oncology. In certain cases, such as the small molecule imatinib in the treatment of chronic myelogenous leukemia and the monoclonal antibody trastuzumab in the treatment of HER2/neu-amplified breast cancer, these targeted therapies have led to significant improvements in clinical outcomes.1, 2 Since the discovery that the v-sis oncogene is derived from the platelet-derived growth factor (PDGF)-B chain gene,3 the PDGF receptor (PDGFR) has been an appealing target for anticancer therapeutic development.
It has been demonstrated that PDGF/PDGFR signaling influences cancer biology through 3 distinct but interacting mechanisms: 1) the autocrine growth stimulation of cancer cells, 2) the regulation of stromal-derived fibroblasts, and 3) the regulation of angiogenesis.4 The specific targeting of PDGFRs with a therapeutic antibody may interrupt this signaling axis, suppressing tumor growth by inhibiting the tumor directly and/or disrupting its supporting stroma and blood vessels. Antibody-based approaches may have some advantages over approaches based on small molecules, including greater target specificity, less off-target toxicity, and an additional capacity for immune-mediated cytotoxicity.5 In this report, we review the PDGF and PDGFR system, focusing on PDGFRα as a target for anticancer therapeutic development. In addition, agents in current use or development that target PDGF/PDGFR, particularly IMC-3G3 (a fully human immunoglobulin G subclass 1 [IgG1] monoclonal antibody to PDGFRα), are reviewed.
The PDGF/PDGFR Axis in Cellular Signaling
The PDGFs are a family of ligands consisting of disulfide-bonded PDGF-A, PDGF-B, PDGF-C, or PDGF-D polypeptide chains.6 The individual ligand monomer chains form the following heterodimers or homodimers: PDGF-AA, PDGF-AB, PDGF-BB, PDGF-CC, and PDGF-DD.7 Ligand dimerization is a requirement for signaling, which occurs when these PDGF dimers bind to and activate 2 structurally related cell surface tyrosine kinase receptors, PDGFRα and PDGFRβ.8
PDGFRα and PDGFRβ both are members of the class III receptor tyrosine kinase (RTK) family and have molecular weights of approximately 170 kilo Daltons (kDa) and 180 kDa, respectively.7 These receptors are the products of 2 homologous genes located on chromosomes 7 and 22, respectively.9 The extracellular region of each receptor contains 5 immunoglobulin (Ig)-like domains, whereas the intracellular region contains a tyrosine kinase domain.10 Upon PDGF binding, receptor homodimerization or heterodimerization is induced, resulting in the formation of PDGFR-αα, PDGFR-αβ, and PDGFR-ββ. All PDGF isoforms except PDGF-DD are capable of inducing homodimerization of PDGFRα, whereas only PDGF-BB and PDGF-DD are capable of inducing homodimerization of PDGFRβ. In addition, in cells that express both PDGFRα and PDGFRβ, heterodimerization of the receptors can be induced by all PDGF isoforms except PDGF-AA (Fig. 1).11
After dimerization, each PDGFR partner phosphorylates the other on specific tyrosine residues located on their cytosolic tails.7 This autophosphorylation activates downstream signaling by 1) further boosting the kinase activity of the receptor through the additional phosphorylation of tyrosine residues within the kinase domain, and 2) creating high-affinity docking sites for the Src homology 2 (SH2) domain-containing adapter molecules (eg, Grb2, Grb7, Crk, Nck, and Shc) that lack intrinsic enzymatic activity but serve as links to other downstream effectors.6
PDGFR-αα, PDGFR-αβ, and PDGFR-ββ transduce a variety of cellular signals that act to prevent apoptosis (survival signaling), stimulate mitogenesis (proliferative signaling), stimulate cellular chemotaxis, and increase the concentration of intracellular calcium.12-14 These effects are mediated through several well characterized downstream signaling cascades induced by the activation of PDGFRs, including the Ras/mitogen-activated protein kinase (MAPK) pathway, the phosphatidylinositol 3-kinase (PI3-K) pathway, and the phospholipase C-γ1 (PLC-γ1) pathway. The PI3-K pathway is critical for PDGF-induced prevention of apoptosis, actin reorganization, and cell migration.15 Ras and its associated downstream MAPK cascade mediate the cell proliferative effects of PDGF. Ras is regulated through the adapter molecule Grb2 in conjunction with son of sevenless-1 (sos-1) and by subsequent modulation of the ratio of the inactive guanosine diphosphate-bound Ras form to the active guanosine triphosphate-bound form. Ras signal transduction also may affect cytoskeletal organization and cell morphology by activating the tyrosine kinases Src and Abl.16 The tyrosine kinase phosphorylation of PLC-γ1 by PDGF leads to an increase in the turnover of phosphatidylinositol,17 a second messenger that increases intracellular calcium stores and thereby regulates calcium-driven intracellular signaling.18
The Role of PDGF in Development and Disease
It has been demonstrated that both PDGFRα and PDGFRβ are involved in organ development and function and in the pathogenesis of several nonmalignant and malignant diseases.7, 19-23 The evidence for a critical role played by PDGF/PDGFR signaling during embryonic development includes the finding that deletion of either receptor is lethal to the embryo.19, 20 PDGFRα is important for the mesenchymal cell development of many tissues, including the intestines, kidney, and skin. The receptor also plays a role in regulating the development and maturation of chondrocytes, non-neuronal neural crest, and lung alveolar smooth muscle.7, 21, 22 PDGFRβ signaling is critical for normal vascular smooth muscle development and pericyte recruitment to capillaries19 and for development of the peripheral nervous system.23
In normal adult tissues, PDGFR expression varies widely among cell types.24 High-density PDGFRα expression is characteristic of platelets, megakaryocytes, fibroblasts, pericytes, vascular smooth muscle cells, neurons, and myoblasts. The predominant cell types with high-density PDGFRβ expression are fibroblasts, pericytes, and smooth muscle cells.4, 7, 24
It has been demonstrated that PDGF signaling affects wound repair as well as the pathogenesis of several nonmalignant diseases. PDGFs accelerate inflammation and wound healing by activating mitogenesis, chemotaxis, and protein synthesis in PDGFR-positive fibroblasts and smooth muscle cells.25-27 Hyperactive signaling along the PDGF/PDGFR axis also has been demonstrated in several nonmalignant disorders, such as atherosclerosis,28, 29 pulmonary fibrosis (both PDGFRα and PDGFRβ),30, 31 hepatic cirrhosis (PDGFRβ only),32, 33 and glomerular diseases characterized by mesangial proliferation.34-40 PDGFRβ-associated functions in cancer biology include pericyte recruitment and the regulation of interstitial fluid pressure in tumors, and the latter may affect the uptake of anticancer therapeutics into malignant tumors.41-43
The Role of PDGFRα in Malignancy
PDGF/PDGFRα signaling plays a relevant role in cancer biology not only because of direct effects on tumor cells but also because of paracrine effects mediated by PDGFRα expressed on tumor stroma and effects mediated through the regulation of malignant angiogenesis (Fig. 2).44-60 Several types of cancers, particularly those derived from the ovary, prostate, breast, lung, brain, skin, and bone, express PDGFRα on the malignant cells themselves (Table 1).22 In 1 experiment in which a human tumor array was probed with a cross-reactive polyclonal rabbit antibody to PDGFRα, PDGFRα expression was noted in approximately 95% of osteosarcomas and chondrosarcomas, in 77% of prostate cancers, in 52% of ovarian cancers, in 65% of breast cancers, and in 51% of lung cancers (N. Loizos, ImClone Systems, unpublished results). A separate investigation demonstrated strong immunohistochemical staining (2 to 3+) in 89% of the clear cell histologic subtype of ovarian cancers.50 Gene amplifications of PDGFRα have been identified in a subset of malignant gliomas,62 and activating mutations of PDGFRα have been discovered in a proportion of gastrointestinal stromal tumors.63 Tumor PDGFRα expression has been associated with disease progression in renal cell cancer,64 diminished survival in ovarian cancer,56 lymph node metastasis in breast cancer,58 and bone metastases in prostate cancer.59
Elevated expression compared with normal tissue using Western blot analysis
Hepatocellular cancer (∼70%): Stock 200748
Glioblastoma multiforme (24%): Fleming 199262
Investigations in multiple cancer types have highlighted the relevance of both PDGF-activated tumor and supporting stromal cells in facilitating cancer growth and metastasis.44-51, 55, 57, 60 In some cases, specific autocrine or paracrine mechanisms that contribute to malignant progression have been suggested. In 1 experiment, NIH 3T3 cells were transformed when PDGF-A or PDGF-B gene expression was induced using a transfected promoter.44 In another preclinical study, an autocrine mechanism of ovarian cancer growth was identified by 1) a strong correlation of immunohistochemical staining for PDGF-AB and PDGFRα on cancer cells, and 2) the finding that growth media conditioned by ovarian cancer cells contained PDGF and, moreover, activated cellular proliferation through PDGFRα expressed on cancer cells.50 Whereas 1 investigation of human melanoma xenografts demonstrated that PDGF-driven paracrine signaling was required for the development of malignant stroma,46 a separate study indicated that the induction of PDGF-CC expression in melanoma cells led to accelerated tumor growth through the activation of PDGFRα.47 In the latter study, the exclusive expression of PDGFRα on cancer-associated stromal fibroblasts illuminated a paracrine mechanism of stromal-regulated melanoma tumor growth. It also has been demonstrated that PDGF/PDGFRα-associated autocrine or paracrine mechanisms that involve tumor and stroma similarly play important roles in the growth of hepatocellular cancers48, 49 and lung cancers,51 in facilitating bone metastasis in prostate cancers,60 and in the metastatic spread of lung and breast cancers.45, 57
It also has been established that an intact PDGF/PDGFRα axis is required for vascular endothelial growth factor (VEGF) production by tumor stroma and for the regulation of malignant angiogenesis in tumors that are resistant to antiangiogenic treatment.52-54 The former was demonstrated in an elegant experiment in which VEGF-null fibrosarcomas grown in mice required the recruitment of PDGFRα-expressing stromal fibroblasts for VEGF production, angiogenesis, and tumorigenesis.52 The recruitment of PDGFRα-expressing stromal fibroblasts depended on paracrine stimulation by PDGF-AA produced from the tumor cells. When the PDGF/PDGFRα axis was disrupted, both angiogenesis and tumorigenesis were reduced significantly. The importance of the PDGF/PDGFRα axis in the regulation of malignant angiogenesis was demonstrated in another study in which tumor-associated fibroblasts (TAFs) from antiangiogenic therapy-resistant lymphomas (EL4) stimulated the growth of antiangiogenic therapy-sensitive lymphomas (TIB6), even in the presence of VEGF inhibition.53 PDGF-C was up-regulated in these TAFs from EL4 (resistant) tumors, and PDGF-C-neutralizing antibodies blocked angiogenesis and impeded cancer growth, suggesting that PDGF-C may be instrumental in mediating resistance to antiangiogenic therapy. The same investigation demonstrated that endothelial cells within EL4 lymphomas expressed both PDGFRα and PDGFRβ, raising the possibility that direct effects on endothelial cells also may explain in part PDGF-mediated malignant angiogenesis. Collectively, these observations suggest that therapeutics targeting the PDGF/PDGFRα pathway may exert anticancer effects not only through the interruption of autocrine-driven and paracrine-driven pathways of tumor growth but also through paracrine effects on cancer-associated vasculature.
Overview of PDGF-Targeting Agents
A partial list of PDGF/PDGFR-targeting agents is provided in Table 2. Although most are small-molecule tyrosine kinase inhibitors that inhibit multiple RTKs, 2 (IMC-3G3: ImClone Systems, Branchburg NJ; MEDI-575: MedImmune, Gaithersburg, Md) are PDGFRα-targeting monoclonal antibodies.
Several small molecules nonspecifically inhibit the PDGF/PDGFR axis, but with less potency compared with their principal effects against other targets. For example, imatinib, a 2-phenylaminopyrimidine–derivative tyrosine kinase inhibitor that principally targets Abl and c-kit, also inhibits the tyrosine kinases of PDGFRα and PDGFRβ.65 Imatinib currently is approved in the United States for the treatment of 1) Philadelphia chromosome-positive chronic myelogenous leukemia and acute lymphoblastic leukemia, both of which have a characteristic bcr-abl translocation and increased kinase activity66, 67; and 2) gastrointestinal stromal tumors, which commonly have activating c-kit mutations.68 Indications for the use of imatinib that arise from imatinib's specific effects on PDGFR include dermatofibrosarcoma protuberans, which has a distinguishing translocation involving PDGF-B,69 and idiopathic hypereosinophilic syndrome, which is associated with interstitial deletions involving PDGFRα.70 To date, clinical trials of imatinib designed to take advantage of its inhibition of PDGFRα expressed in several types of malignancies, including ovarian cancer,71-74 lung cancer,75 prostate cancer,76 and glioblastoma multiforme,77-78 have demonstrated no activity or limited activity.
Development, Preclinical Results, and Preliminary Clinical Results With IMC-3G3 (Human Anti-PDGFRα IgG1 Monoclonal Antibody)
In light of the considerable role played by PDGFRα in cancer growth, survival, and metastasis, the fully human IgG1-type anti-PDGFRα monoclonal antibody IMC-3G3 was developed as an anticancer therapeutic agent. Hybridomas that produced high-affinity monoclonal antibodies to the extracellular region of PDGFRα were generated using human IgG transgenic mice.79 Further selection led to the identification of IMC-3G3 as the most potent and specific PDGFRα-neutralizing antibody. IMC-3G3 subsequently was cloned into suitable vectors for sequence analysis and stable expression as full-length IgG1 in mammalian cells. IMC-3G3 binds to PDGFRα with high affinity (Kd, 0.04 nM) and inhibits ligand binding with a 50% inhibitory concentration (IC50) of 0.24 nM to 0.58 nM.79 The antibody does not cross-react with PDGFRβ or with murine PDGFRα.79 IMC-3G3 is approximately 100-fold more potent than imatinib with respect to the inhibition of PDGF-induced cell proliferation (IC50, 1 nM for IMC-3G3 vs 100 nM for imatinib) and PDGF-mediated PDGFRα signaling (IC50, 10 nM for IMC-3G3 vs 1 μM for imatinib) (N. Loizos, ImClone Systems, unpublished results).
Tumor stromal cells and stromal-influenced angiogenic factors are important for facilitating PDGF/PDGFRα-mediated effects on cancer growth, as discussed above. Both in vitro and in vivo experiments, in which human cancer cells or animals bearing human tumor xenografts are treated with IMC-3G3, are likely to underestimate its potential anticancer activity, because cell lines contain neither stroma nor blood vessels, and IMC-3G3 cannot react with mouse-derived stroma or regulate cancer-associated vasculature in xenograft experiments. In humans, IMC-3G3 is expected to target PDGFRα expressed in both tumor and stroma, thereby potentially inhibiting tumor, stroma, and vasculature.
Preclinical studies of IMC-3G3 as a single agent or in combination with cytotoxic chemotherapy have demonstrated antimitogenic activity in cancer cell lines as well as antitumor growth activity in human xenografts. One set of in vitro experiments demonstrated that IMC-3G3 was capable of inhibiting any PDGF-induced proliferation or signaling through PDGFRα in both normal cell lines and cancer cell lines, including U118 (glioblastoma) and SKLMS-1 (leiomyosarcoma) cells.79 In a separate study of mouse and human hepatoma cell lines grown to confluence, treatment with IMC-3G3 as a single agent resulted in significant decreases in both cell proliferation and survival.48 Similarly, IMC-3G3 monotherapy led to a significant decrease in PDGFRα-driven cell proliferation in ovarian cancer cell lines.50 In vivo studies demonstrated that treatment with IMC-3G3 as a single agent inhibited the growth of glioblastoma (U118; 65% tumor growth inhibition) and leiomyosarcoma (SKLMS-1; 69% tumor growth inhibition) human xenografts (Fig. 3), both of which were chosen because of their expression of PDGFRα.80 In the case of the glioblastoma xenografts, IMC-3G3 monotherapy also reduced both PDGFRα activation79 and activation of the MAPK proliferation and Akt survival pathways (N. Loizos, ImClone Systems, unpublished results). Finally, a combination study of IMC-3G3 and doxorubicin conferred greater tumor growth inhibition than either IMC-3G3 or doxorubicin monotherapy in KHOS/NP human osteosarcoma cancer xenografts (unpublished data).
On the basis of this preclinical activity and the absence of adverse effects associated with IMC-3G3 when administered to cynomolgus monkeys at doses of up to 75 mg/kg,80 IMC-3G3 entered clinical development for patients with advanced solid malignancies. In a phase 1 study of patients with advanced solid tumors and lymphomas who failed standard therapy or for whom no standard therapy was available, patients received IMC-3G3 intravenously as a single agent during the initial 6 weeks of therapy (4 weekly or 2 biweekly infusions followed by a 2-week observation period).80 Patients who experienced an objective response or stable disease continued to receive additional therapy at the same dose and schedule in the absence of disease progression or other withdrawal criteria. A preliminary review of the study results, at which time 20 patients were enrolled in 5 dose cohorts (4 mg/kg, 8 mg/kg, and 16 mg/kg weekly; and 15 mg/kg and 20 mg/kg biweekly), revealed that IMC-3G3 was well tolerated, and there were no specific drug-related adverse events reported.80 One patient with prostate cancer that was refractory to hormone therapy and chemotherapy experienced a prostate-specific antigen decrease >50% after treatment with IMC-3G3 at the 4-mg/kg dose level.80 Interim pharmacokinetic results indicated that weekly infusions of 16 mg/kg and biweekly IMC-3G3 infusions of 15 mg/kg or 20 mg/kg could attain or exceed the trough concentrations required for human tumor growth inhibition in xenograft studies.80 On the basis of these preclinical and early clinical findings, several phase 2 clinical trials of IMC-3G3 in advanced solid malignancies are currently in development.
Summary and Conclusions
PDGFRα inhibition by IMC-3G3 may impede cancer growth through the direct inhibition of cancer cells and supportive stroma and through the regulation of malignant angiogenesis. Preclinical evaluations with IMC-3G3 have demonstrated that the antibody inhibits cancer growth in selected cancer cell lines and human xenograft models and confers enhanced anticancer activity when combined with doxorubicin. However, it should be recognized that the activity of IMC-3G3 in humans may not be predicted adequately by current preclinical models because of the absence of stroma and blood vessels in cell lines and because of the lack of cross-reactivity of IMC-3G3 with murine PDGFRα, which is expressed on mouse-derived stroma in xenograft models. Preliminary clinical results with IMC-3G3 reveal a favorable safety profile. In the future, combination studies of PDGFR-targeting therapeutics with antiangiogenic agents may take further advantage of the potential interplay among tumor, stromal, and angiogenic factors that appear to be a hallmark of PDGFR signaling.
CONFLICT OF INTEREST DISCLOSURES
The papers in this supplement represent proceedings of the “12th Conference on Cancer Therapy with Antibodies and Immunoconjugates,” in Parsippany, New Jersey, October 16-18, 2008. Unrestricted grant support for the conference was provided by Actinium Pharmaceuticals, Inc.; Bayer Schering Pharma; Center for Molecular Medicine and Immunology; ImClone Systems Corporation; MDS Nordion; National Cancer Institute; National Institutes of Health; New Jersey Commission on Cancer Research; and PerkinElmer Life & Analytical Sciences. The supplement was supported by an unrestricted educational grant from ImClone Systems Corporation, a wholly owned subsidiary of Eli Lilly and Company, and by page charges to the authors. The authors are full-time employees of ImClone Systems, a wholly owned subsidiary of Eli Lilly and Company.