Funded Grants


Mechanisms and inhibition of anti-angiogenic therapy-induced glioma invasion

There are two distinct hallmarks of malignant diffuse glioblastoma that are well documented and recognized by physicians and pathologists: the insidious tumor infiltration of single cells into the brain parenchyma far away from the primary tumor site and the robust tumor angiogenesis. The first feature of these deadly tumors presents a continuous challenge for complete surgical resection of glioblastoma and a difficult task for radiation and chemotherapies. The latter refers to the formation of blood vessels in high-grade glioblastoma that allows rapid glioma tumor growth and aggressive tumor invasion in the brain. Disappointedly, despite rapid advance in treatment by surgical removal, radiation, chemo- and immuno-therapies, the median survival time of patients with malignant glioblastoma remains 12 to 16 months. This clinical challenge underscores the importance of better understanding of mechanisms underlying these two malignant features of malignant glioblastomas and developing novel therapeutic regiments to treat these deadly brain cancers.

The major function of glioma angiogenesis is to supply nutrients, oxygen, various necessary substances to the tumors, and remove metabolic wastes from the tumors. In theory, inhibition of tumor angiogenesis by cutting off the nourishment and blocking the waste disposal will starve tumors to death, thus effectively shrinking malignant glioblastoma. Consequently, anti-angiogenic treatments that target angiogenic stimulators such as vascular endothelial growth factor (VEGF) produced by glioma cells and VEGF receptor (VEGFR) in tumor endothelial cells at the inner lining of blood vessels are used with increasing frequency to treat patients with glioblastoma (1). Following its FDA approval as monotherapy for recurrent high-grade glioblastoma multiforme (GBM) in 2009, Bevacizumab (Bev), a VEGFspecific monoclonal antibody, has entered mainstream clinical treatments for malignant GBMs. However, various phase II clinical trials reported that treatment of glioblastoma patients with Bev in combination with chemotherapeutic agents showed brief initial response. Afterwards, almost 100% patients who received Bev inevitably developed even more malignant phenotype; glioblastoma has become more aggressively invasive in the brain. As a result patients quickly succumb to the disease. Furthermore, these clinical observations have been corroborated by two recent preclinical studies using various tumor xenograft models including gliomas in animals.

Our group has a long-standing interest in studying the mechanisms of glioma angiogenesis and invasion, and exploring therapeutic approaches to overcome these two malignant phenotypes of glioblastomas. In our studies of glioma angiogenesis and invasion, we reported that the growth factor angiopoietin-2 (Ang2) is capable of stimulating malignant glioma invasion in the brain as well as promoting breast cancer metastasis to distant organs of animals (2). Additionally, Dr. S. Ferrone, a co-investigator of this study had studied chondroitin sulfate proteoglycan 4 (CSPG4) in other types of human cancers (3). CSPG4 has been demonstrated to promote glioma cell invasion (4). In our preliminary studies, we established a glioma model that mimics this malignant phenotype in the brain of animals. When glioma-derived VEGF is trapped by a soluble VEGFR, thus lost its function, VEGFstimulated endothelial cell migration is inhibited in vitro. In the brain, inhibition of gliomaderived VEGF by this trap induces aggressive glioma invasion in the brain of mice. We found that expression of Ang2 and CSPG4 are increased in invasive sVEGFR gliomas compared with controls. In peripheral blood of mice bearing these invasive brain gliomas, serum levels of these two proteins are increased compared to mice bearing non-invasive brain tumors. Significantly, serum levels of Ang2 and CSPG4 are also elevated in serum samples from glioblastoma patients treated with Bev compared to normal individuals.

Our overall hypothesis of this study is that Bev inhibition of VEGF signaling in GBMs induces tumor cell invasion through up-regulation of Ang2 and CSPG4 activating unique signaling pathways. Therefore, simultaneously targeting of VEGF signaling and of the evoked glioma invasion should overcome this malignant phenotype and be effective in treatment of diffuse glioblastomas.

We will test this hypothesis by pursuing the following questions: 1) What are the mechanisms by which anti-VEGF therapy up-regulates Ang2 and CSPG4 inducing glioma cell invasion? 2) Can we inhibit anti-VEGF-induced glioma invasion by targeting up-regulated Ang2, CSPG4 and activated signaling? 3) Can we establish Ang2 and CSPG4 as well as other molecules as biomarkers predictive of anti-angiogenic therapy-induced glioma invasion? To address these questions, we propose three specific aims in this study.

In our first specific aim, we will identify the key modulators that up-regulate Ang2 and CSPG4 in response to anti-angiogenesis therapy in glioma cells. Previous studies showed that expression of Ang2 in endothelial cells, a cervical cancer cell line called HeLa cells and adipose tissues could be induced by hypoxia, i.e. low oxygen condition that frequently occurs in rapidly-growing gliomas and also can be caused by inhibition of VEGF and angiogenesis through several molecules such as HIF1-α, ETS-1 and FOXC2 that control gene transcription. We will first examine whether hypoxic condition induces Ang2 and CSPG4 in glioma cells when VEGF is inhibited and whether other types of brain cells are also involved. Then, we will separately remove these three genes from the glioma cells using siRNA knockdown technology and test whether hypoxia and brain cells still capable of inducing Ang2 and CSPG4 in glioma cells. This aim will identify the modulators of Ang2 and CSPG4 expression that are responsible for anti-VEGF-induced Ang2 and CSPG4 and the induced glioma invasion.

In our second specific aim, we will perform proof-of-principle preclinical studies to explore therapeutic approaches for overcoming the anti-angiogenesis therapy-induced glioma invasion in the brain. We will use siRNAs to deplete Ang2 and CSPG4 in glioma cells that glioma-derived VEGF is trapped and in clinical relevant primary glioblastoma cells that maintain clinical features of tumor invasion and gene alterations. We will assess the impact of Ang2 and CSPG4 depletion on anti-angiogenesis therapy-induced glioma invasion in the brain. Next, we will separately deplete HIF1-α, ETS-1 and FOXC2 in these cells and examine the impact on glioma invasion. Lastly, we will use a 7T magnet resonance (MR) imager designed for small animals installed at our animal facility. With help of coinvestigators who are experts of MR imaging, we will employ MRI analyses to evaluate antiangiogenesis therapy-induced glioma invasion and glioma responses to the combined inhibition in the brain of animals. This aim will identify Ang2, CSPG4 and other molecules as therapeutic targets for combined therapies with anti-angiogenesis drugs and chemotherapeutic agents to treat the malignant glioblastomas and obtain critical MRI data useful for clinical MRI analyses and diagnosis.

In our third specific aim, we will determine whether an increase in serum levels of Ang2 and CSPG4 in the peripheral blood of glioma patients treated with Bev correlates with disease progression. Accumulated clinical data shows that elevated serum levels of Ang2 and CSPG4 are closely linked to tumor progression, invasion and metastasis and patient survival in various types of human cancers including glioblastomas (2, 3). In collaboration with a neuro-oncologist/physician-scientist, Dr. H. Okada, we will prospectively collect serum samples from glioblastoma patients who will be treated with Bev after their diagnosis of glioblastoma. We will determine the serum levels of Ang2, CSPG4 as well as other proteins in sera of these patients after the diagnosis and onward during the Bev treatments. We will also compare the serum levels of these proteins of Bev-treated glioma patients with normal individuals. Lastly, with help of the Biostatistics Core at our NCI-designated comprehensive Cancer Center, the University of Pittsburgh Cancer Institute, we will correlate serum levels of these proteins that changed in response to anti-angiogenesis therapy with disease progression of glioblastomas in response to anti-angiogenesis therapy. This aim will establish Ang2, CSPG4 as well as several other proteins as robust biomarkers predictive for anti-angiogenic therapy-induced glioma invasion during the treatment of glioblastomas.

In clinic, the malignant phenotype of anti-angiogenesis therapy-induced glioma invasion remains a major obstacle and a challenging problem for which no effective salvaging therapy has identified. Based on our novel findings and functions of Ang2 and CSPG4 in inducing glioma invasion and tumor metastasis, we propose to investigate the role and mechanism of Ang2 and CSPG4 in promoting anti-angiogenesis therapy-induced glioblastoma invasion. This study is highly innovative with significant clinical impacts since we will not only elucidate the novel role of Ang2 and CSPG4 in anti-angiogenesis therapy-induced glioblastoma invasion and the underlying mechanisms, identify Ang2, CSPG4 and their modulators as potential therapeutic targets, but we will also establish Ang2, CSPG4 and other molecules as prognostic biomarkers predictive for this adverse phenotype. Additionally, this proposal draws the unique strength of a highly collaborative research team comprised of several established investigators with distinct expertise in immunotherapy for human cancers and glioblastomas, MRI analysis of human cancers and a physician-scientist, thus warranting its likely success. Completion of this study will add elucidation to pathophysiology of the resistance to anti-angiogenic treatment, allow us to explore therapeutic approaches to inhibit the anti-angiogenesis therapy-induced glioma invasion and provide information to monitor glioma progression in response to Bev treatment and implement different therapies in these patients in a timely fashion.

REFERENCES:

  1. Norden AD, Drappatz J, Wen PY. Antiangiogenic therapies for high-grade glioma. Nat Rev Neurol 2009; 5: 610-20.
  2. Hu B, Cheng SY. Angiopoietin-2: development of inhibitors for cancer therapy. Curr Oncol Rep 2009; 11: 111-6.
  3. Wilson BS, Imai K, Natali PG, Ferrone S. Distribution and molecular characterization of a cell-surface and a cytoplasmic antigen detectable in human melanoma cells with monoclonal antibodies. Int J Cancer 1981; 28: 293-300.
  4. Stallcup WB, Huang FJ. A role for the NG2 proteoglycan in glioma progression. Cell Adh Migr 2008; 2: 192-201.