Angiogenesis In Brain Tumor Models

Marked neovascularization is a hallmark of many neoplasms in the CNS. Vascular pathology is a key feature of glioblastoma multiforme characterized by hypervascularity, vascular permeability and hyper-coagulability. Vascular endothelial growth factor (VEGF) has been investigated as a potent mediator of brain tumor angiogenesis, vascular permeability, and glioma growth and is known to be upregulated in most cases of glioblastoma multiforme [89-101]. Microvessel density and VEGF levels have been shown to be independent prognostic markers of survival in fibrillary low-grade astrocytoma. Tumors with a larger number of microvessels also had a greater probability of undergoing malignant transformation [90]. Another study examined the activated phosphorylated form of the KDR receptor in astrocytic neoplasms and found the phosphorylated form of KDR in fresh surgical specimens of glioblastomas (71 per cent) and anaplastic gliomas (15 per cent), but not in low grade gliomas indicating that onset of angiogenesis is an important event during the disease progression of gliomas [96]. Chan et al., [100] found the VEGF receptors, KDR and Flt-1 to be upregulated in the tumor vasculature of glioblastoma multiforme, anaplastic oligodendroglio-mas, and ependymomas with necrosis but not in astrocytomas grade II, anaplastic astrocytomas, or oligodendroglioma tumors. In meningiomas, VEGF was associated both with tumor vascularity and peri-tumoral edema [91]. In looking for a correlation between angiographic neovascularization, peri-tumoral brain edema and the expression of vascular endothelial growth factor, Bitzer et al., [95] found that tumors with high VEGF staining had a significantly higher edema index and a higher edema incidence. In addition, all of the meningiomas with very high VEGF expression were associated with vascular tumor supply from cerebral arteries. Takano et al., [93] found that VEGF concentrations of glioblastoma cyst fluid were 200- to 300-fold higher than those of serum in the patients. VEGF concentration in the tumors was significantly correlated with the vascu-larity measured by counting vessels stained with von Willebrand factor antibody. VEGF is expressed in a wide spectrum of brain tumors and is associated with neovascularization.

However, other angiogenic factors also appear to contribute to the vascularization of CNS neoplasms [92]. The expression of angiopoietin-1 and angiopoietin-2 in human astrocytomas was investigated by in situ hybridization [99]. Angiopoietin-1 mRNA was localized in tumor cells and angiopoietin-2 mRNA was detected in endothelial cells. The results suggested that angiopoietins are involved in the early stage of vascular activation and in advanced angiogenesis and indicate that angiopoietin-2 may be an early marker of glioma-induced neovascularization. Takano et al., [102] investigated the expression of the angiogenic factor thymidine phosphorylase in human astrocytic tumors and found that thymidine phosphorylase was expressed in the tumor cells, macrophages, and endothelial cells. The influence of antiangiogenic treatment on 9L gliosarcoma oxygenation and response to cytotoxic therapy was investigated [103]. In another study, the mean concentrations of VEGF were found to be 11-fold higher in high grade gliomas and the mean concentrations of hepatocyte growth factor/scatter factor (HGF/SF) were found to be 7-fold higher in high grade gliomas than in low grade tumors [104]. In addition, VEGF and HGF/SF appeared to be independent predictive parameters for glioma microvessel density. The findings of this study also suggested that basic fibroblast growth factor (bFGF) is an essential cofactor for angiogenesis in gliomas.

The signal transduction from extracellular protein growth factors occurs by a variety of mechanisms that share many common features. Activation of specific receptor kinases do not activate unique intracellular kinases which then result in a linear signaling pathway; rather multiple signaling cascades can be activated producing combinatorial effects that allow more refined regulation of the biological outcome [105]. The intracellular signal transduction pathways for VEGF and bFGF in endothelial cells have not been fully elucidated; however, it is likely that protein kinase C is an important pathway component for both mitogens [106-114]. Neoangiogenesis in the eyes of rats bearing corneal micropocket implants of either VEGF or bFGF was inhibited by treatment of the animals with LY317615 orally twice per day [115]. Treatment of human SW-2 small cell lung carcinoma-bearing mice with LY317615 orally twice per day resulted in a dose-dependent decrease in the number of countable intratumoral vessels in the tumors. The number of intratumoral vessels stained by Factor VIII was decreased to one-half of the controls in animals treated with LY317615 and the number of vessels stained by CD31 was decreased to one-quarter of the controls in animals treated with LY317615 [115]. The effects of LY317615 have been explored in a variety of tumor models [116-120].

The human T98G glioblastoma multiforme line was used as a brain tumor model [121]. The protein kinase Cp inhibitor 317615 was not very cytotoxic toward T98G cells in culture and was additive in cytotoxicity with BCNU. When nude mice bearing subcutaneous T98G tumors were treated with LY317615 orally twice daily on days 14 through 30 post-implant, the number of intratumoral vessel stained by CD31 was decreased to 37 per cent of control and vessels stained by CD105 was decreased to 50 per cent of control. The compound LY317615 was an active antitumor agent against subcutaneously growing T98G xenografts (Fig. 4.1) [122]. A treatment regimen administering LY317615 prior to, during and after BCNU was compared with a treatment regimen administering LY317615 sequentially after BCNU. In the tumor growth delay determination of the subcutaneous tumor, the sequential treatment regimen was more effect than the simultaneous treatment regimen (Figs. 4.1 and 4.2). However, when the same treatments were administered to animals bearing intra-cranial T98G tumors, the survival of animals receiving the simultaneous treatment regimen increased from 41 days for BCNU alone to 102 days for the

FIGURE 4.1 Growth delay of subcutaneously implanted human T98G glioblastoma multiforme after treatment with LY317615 (10 or 30 mg/kg) orally twice per day on days 4 through 18 alone or along with BCNU (15 mg/kg, ip) on days 7 through 11, or with SU5416 (25 mg/kg, ip). Points are the means of 5 animals; bars are SEM.

PKCb INHIBITOR DOSE, mg/kg PKCb INHIBITOR DOSE, mg/kg

• . 317615 2HCI: O, LY333531; SU5416 PKCß INHIBITOR or SU5416, d4-18/ BCNU, d7-11

FIGURE 4.1 Growth delay of subcutaneously implanted human T98G glioblastoma multiforme after treatment with LY317615 (10 or 30 mg/kg) orally twice per day on days 4 through 18 alone or along with BCNU (15 mg/kg, ip) on days 7 through 11, or with SU5416 (25 mg/kg, ip). Points are the means of 5 animals; bars are SEM.

RESPONSE OF HUMAN T98G GLIOBLASTOMA MULTIFORME TO SEQUENTIAL TREATMENT WITH A PKCp INHIBITOR AFTER BCNU

PKCb INHIBITOR DOSE, mg/kg PKCb INHIBITOR DOSE, mg/kg

• , 317615 2HCI; O, LY333531 BCNU,d7-11 -> PKCß INHIBITOR, 2x d12-30

FIGURE 4.2 Growth delay of subcutaneously implanted human T98G glioblastoma multiforme after treatment with LY317615 (3, 10, or 30 mg/kg) orally twice per day on days 12 through 30 alone or after administration of BCNU (15 mg/kg, ip) on days 7 through 11, or with LY333531 (50 mg/kg, ip). Points are the means of 5 animals; bars are SEM.

FIGURE 4.3 Survival of animals bearing intracranial human T98G glioblastoma multiforme after treatment with LY317615 (10 or 30 mg/kg) orally twice per day on days 4 through 18 alone or along with BCNU (15 mg/kg, ip) on days 7 through 11, or with SU5416 (25 mg/kg, ip). Data are the means of 5 animals.

FIGURE 4.3 Survival of animals bearing intracranial human T98G glioblastoma multiforme after treatment with LY317615 (10 or 30 mg/kg) orally twice per day on days 4 through 18 alone or along with BCNU (15 mg/kg, ip) on days 7 through 11, or with SU5416 (25 mg/kg, ip). Data are the means of 5 animals.

combination while animals receiving the sequential treatment regimen survived 74 days (Figs. 4.3 and 4.4). Treatment with the protein kinase CP inhibitor decreased T98G glioblastoma multiforme angiogenesis and improved treatment outcome with BCNU [122].

The potential of antiangiogenic agents to augment the antitumor activity of standard cytotoxic chemo-therapeutic agents is becoming well established [123, 124]. Among the antiangiogenic agents under investigation, TNP-470, an inhibitor of endothelial cell proliferation, has been shown to delay the growth of gliomas and other brain tumors in several studies [124, 125]. The antiangiogenic combination of TNP-470 and minocycline increased the response of both intracranial and subcutaneous rat 9L gliosarcoma to BCNU or adriamycin [103]. Lund et al., [125] found that TNP-470 treatment increased the response of subcutaneous human U87 glioblastoma xenografts to radiation therapy but did not increase the response of intracranial U87 to radiation therapy. Angiostatin, an antiangiogenic internal fragment of plasminogen, has been shown to suppress the growth of rat C6 and rat 9L gliomas as well as human U87 glioma whether implanted subcutaneously or intracranially [126]. Angiostatin used in combination with fractionated radiation therapy had a greater-than-additive effect on the growth of a human glioma in nude mice [127]. The PKCP inhibitor LY317615 is an orally administered small molecule without toxicity in rodents at the antiangiogenic doses [115]. The T98G glioblastoma multiforme tumor model allowed the comparison of combination treatment regimens that examined the efficacy of the LY317615 given simultaneously or sequentially with BCNU and the examination of two experimental endpoints, tumor growth delay and survival [122]. The cell culture studies indicate there can be an interaction between PKC inhibition and BCNU to enhance cytotoxicity in the malignant cells. Interestingly, the sequential treatment regimen, giving the protein kinase CP inhibitor after the cytotoxic agent resulted in a greater effect on tumor growth delay but a lesser effect on survival although these differences did not reach statistical significance. There are several possible reasons for the difference. The first possibility is that the sequential regimen involved treatment of a larger tumor burden allowing the impact on subcutaneous tumor growth to be manifest, but the larger tumor burden in the cranium impacted negatively on survival. A second possibility is that the angiogenic factors operative in the subcutaneous

LIFE-SPAN, Days

FIGURE 4.4 Survival of animals bearing intracranial human T98G glioblastoma multiforme after treatment with LY317615 (3, 10, or 30 mg/kg) orally twice per day on days 12 through 30 alone or after administration of BCNU (15 mg/kg, ip) on days 7 through 11, or LY333531 (50 mg/kg, ip). Data are the means of 5 animals.

LIFE-SPAN, Days

FIGURE 4.4 Survival of animals bearing intracranial human T98G glioblastoma multiforme after treatment with LY317615 (3, 10, or 30 mg/kg) orally twice per day on days 12 through 30 alone or after administration of BCNU (15 mg/kg, ip) on days 7 through 11, or LY333531 (50 mg/kg, ip). Data are the means of 5 animals.

tumor are different than the angiogenic factors expressed in the intracranial tumor and that the protein kinase CP inhibitor is more effective against the subcutaneous tumor. It has often been noted that chemotherapeutic agents have very different levels of efficacy depending upon the organ environment of the tumor [128-130].

A normal volunteer and a phase I trial in solid tumor patients demonstrated the drug was very well tolerated at doses that achieve a biologically active serum concentration [131]. Based on the dependence of glioma growth on VEGF-mediated angiogenesis, and the promising preclinical and clinical data, Fine et al., [132] initiated a phase II trial of LY317615 in patients with recurrent and progressive high grade gliomas following standard therapy. Treatment consisted of oral LY317615 administered daily on an every 6-week cycle after which patients underwent a complete physical/neurological, biochemical, and radiographic reevaluation. Patients were stratified based on those taking enzyme-inducing antiepileptic drugs (EIAED; Group B) and those not taking EIAED (Group A) and conducted pharmacokinetic studies. To date, 32 patients (17 patients in Group A and 15 patients in Group B) have been accrued to the trial and 28 patients were evaluable for response. Treatment was well tolerated with only one possible case of drug-related toxicity > grade 1 (Grade 2 thrombocytopenia). Eleven patients have received more than one cycle of treatment (6 patients in Group A and 5 patients in Group B) and several patients have been stable on treatment for greater then 3 months and a number of other patients continue treatment with LY317615. Objective radiographic responses have been seen in 5 patients. LY317615 appears to have antitumor activity against recurrent malignant gliomas [132].

0 0

Post a comment