Introduction

Integrins are heterodimeric transmembrane cell surface receptors that play a key role in the crosstalk between the cell and its surrounding stroma [1]. Twenty-four different integrins have been identified to date. Integrins link the cytoskeleton to the extracellular matrix, and are recognized to be key regulators of tissue structure. Integrins regulate cell adhesion, migration, differentiation, proliferation, and survival during physiological and pathological conditions, including inflammation and cancer. Upon ligation to extracellular ligands (i.e., matrix proteins such as collagens, laminins, vitronectins and fibronectins), integrins activate downstream signaling pathways in concert with growth factor receptors, including platelet-derived growth factor receptor (PDGFR), epidermal growth factor receptor (EGFR) and vascular endothelial growth factor receptor (VEGFR).

Preclinical data indicate that integrins play a key role in cancer initiation and progression [2]. They provide adhesive, migratory, and survival cues to tumor cells and to cells of the tumor microenvironment, including angiogenic endothelial cells. The integrins alphaVbeta3 and alphaVbeta5, among others, are highly expressed not only on the tumor vasculature and angiogenic endothelial cells, but also on tumor cells, including gliomas (reviewed in [3, 4]). Consequently, integrins have been considered as a promising therapeutic target in cancer [5]. Monoclonal antibodies and peptide-based integrin inhibitors are being investigated for their potential therapeutic activity in various tumor types. This strategy is in advanced stage clinical development in glioblastoma, a highly vascular primary brain tumor. The only integrin inhibitor being studied in glioma is cilengitide, we therefore focus this short review on this agent.

Preclinical data on integrins in glioma

In high-grade glioma, and in particular in glioblastoma, overexpression of alphaVbeta3 integrin is well documented. Importantly, alphaVbeta3 integrin is expressed both on angiogenic endothelial cells and on tumor cells [6–8]. Molecular imaging of alphaVbeta3 expression using the tracer [18F]Galacto-RGD and validation by immunohistochemistry revealed that alphaVbeta3 integrin expression was mainly confined to the tumor region and was absent in normal tissue [9]. Selective alphaVbeta3 upregulation in malignant gliomas suggests that integrin signaling in glioblastoma has important functions. Emerging evidence indicates that integrins promote glioblastoma adhesion, migration and angiogenesis [10, 11]. In an autocrine loop hypoxia will recruit alphaVbeta3 and alphaVbeta5 integrins and activate focal adhesion kinase (FAK). Integrin inhibition decreases hypoxia-inducible factor 1alpha (HIF-1alpha) and reduces thus tumor hypoxia, which may lead to increased radiation sensitivity [12].

The integrin alpha6beta1 plays an important role for the regulation of glioma-initiating cells in the perivascular niche [13]. This integrin mediates the interaction of glioma-initiating cells to laminin, an extracellular matrix protein expressed in basement membranes, including those supporting endothelial cells. This interaction provides an anchorage for glioma-initiating cells within the perivascular niche and supports their tumorigenic potential.

Given the role of integrins in promoting glioma growth, invasion and angiogenesis, integrin inhibitors might be ideal therapeutic tools with synergistic activities in conjunction with already established therapeutic modalities, i.e., radiation therapy and cytotoxic chemotherapy. Importantly, because of their antiangiogenic effects [14], integrin inhibitors are expected to normalize the tumor vasculature, which in turn might enhance the effects of radiation therapy and chemotherapy. Indeed the efficacy of these two therapeutic modalities depends on an effective blood perfusion for oxygenation of the tumor (i.e., an important condition for radiation-induced tumor cell killing) and optimal drug delivery. Inhibition of alphaVbeta3 and alphaVbeta5 integrins in hypoxic glioma cells by siRNA decreases HIF-1alpha by regulating FAK. Silencing this signaling cascade in established xenografts indeed reduced hypoxia and decreased angiogenesis [12].

Interestingly, ionizing radiation induces ceramide-mediated apoptosis of tumor endothelial cells, causing tumor vessel disruption and delayed tumor growth [15]. Consistent with a cytotoxic effect of radiotherapy on angiogenic vessels, concomitant administration of antiangiogenic drugs decreases endothelial cell survival. Anti-VEGF antibodies (e.g., bevacizumab) [16], VEGFR inhibitors (e.g., AG-013736) [17], mTOR inhibitors (e.g., RAD001) [18], and integrin inhibition (cilengitide) [19] sensitize angiogenic endothelial cells to ionizing radiation-induced death, thereby enhancing tumor vascular damage induced by radiotherapy and improving therapeutic response [20]. Irradiation of glioma cells increases their expression of alphaVbeta3 integrin [21], thus integrin inhibition will further synergize with radiation therapy.

Regarding clinical translation, targeting alphavbeta3 and alphavbeta5 might reduce hypoxia and thus resistance to radio- and/or chemotherapy [12]. Indeed, arresting integrin signaling synergistically enhances the anti-glioma effect of radiation therapy in vivo [22].

A recent preclinical study suggests that low concentrations of small molecular integrin inhibitors used as single agents may stimulate angiogenesis and tumor growth in experimental melanoma and Lewis Lung carcinoma models, through increased VEGF expression and VEGFR2 signaling [23]. The concentrations of the small molecular integrin inhibitor cilengitide, measured in patients in clinical trials, however, largely exceed the critical “proangiogenic” concentrations described in this preclinical study, and therefore, paradoxical proangiogenic effects of small molecular integrin inhibitors in the clinical application seem rather unlikely [24]. In fact, relative cerebral blood volume and blood flow, measured by perfusion MRI, remained stable or decreased among recurrent GBM patients responding to cilengitide therapy in a prior phase I study [25]. Durable and clinically relevant responses have been observed with cilengitide monotherapy; ongoing clinical studies administer cilengitide in combination with radiotherapy and/or chemotherapy. Nevertheless, the durable therapeutic success will depend on the efficient management of escape mechanisms occurring under anti-integrin therapies [26, 27].

Integrin inhibition attenuates invasion and metastasis secondary to radiation-induced angiogenesis inhibition

Vascular normalization may not be the only mechanism by which integrin inhibitors synergize with radio-chemotherapy. We have recently observed that high doses of ionizing radiation locally suppress angiogenesis in vivo and inhibited endothelial cell sprouting in vitro through a cell-autonomous effect mediated by the TGFbeta receptor ALK5 [28]. Tumors derived from oral squamous cell carcinoma and from hepatoma cell lines growing within a pre-irradiated bed had decreased microvascular density and were smaller, but their margins were highly invasive. They also had increased hypoxia and necrosis, and a more aggressive behavior with a higher number of lung metastases [29]. Thus, pre-irradiation of the tumor bed reduces local tumor growth, likely due to acute metabolic starvation secondary to suppressed angiogenesis, but at the same time it may enhance local invasion and metastatic spreading, possibly through increased hypoxia. We have identified the cysteine-rich protein 61 (CYR61), a member of the CCN (CYR61/CTGF/NOV) family of matricelluar proteins regulating cell growth, differentiation, survival and migration in development, tissue remodeling and repair [30], and integrin alphaVbeta5 (a receptor for CYR61) expressed by tumor cells, as critical molecules that cooperate to promote local invasion and distant metastases [29]. Importantly, function-blocking anti-αV mAb 17E6 [29, 31] and cilengitide-inhibited CYR61-mediated invasion and metastasis formation of tumor cells growing in a pre-irradiated bed. These results suggest that cilengitide may improve cancer control in conjunction with radiotherapy by attenuating some of the unwanted rebound effect of radiation-induced inhibition of angiogenesis. Hypoxia is well known to activate pro-invasive programs [32] and select for resistant variants with invasive characteristics. This view is consistent with a recent experimental study demonstrating that selected inhibition of tumor angiogenesis by antiangiogenic drugs, including in experimental glioma, results in increased invasion and metastasis formation [33].

Cilengitide

Cilengitide (Merck KGaA, Darmstadt, Germany) is a synthetic Arg-Gly-Asp (RGD) pentapeptide recognizing the RGD ligand-binding motif (ligand binding site) on the integrin receptors alphaVbeta3 and alphaVbeta5 [34, 35] and competitively blocks integrin ligand binding. It was shown to diminish angiogenesis in vitro [36]. In an important early preclinical study, cilengitide markedly suppressed tumor growth in a medulloblastoma and orthotopic glioblastoma models (i.e., tumors were grown in the brain), while no growth inhibition was demonstrated in a heterotopic model (i.e., when tumors were grown in the flank of nude mice), or when an inactive peptide was used [37]. This suggests that the brain environment is particularly susceptible to integrin inhibition and led to subsequent clinical investigation.

In phase I studies cilengitide was administered twice weekly by intravenous infusion over 1 h. No dose limiting toxicity was observed with doses up to 2400 mg/m2. Peak plasma concentrations that had shown antitumor effects in preclinical models were achieved at doses ≥120 mg/m2 [25, 38–40]. The terminal half-life of cilengitide is 3–5 h, suggesting the possibility of increased efficacy with a more frequent administration or continuous infusion schedule.

Clinical experience with cilengitide in malignant glioma

Sustained responses in recurrent glioblastoma were seen both with lower and higher doses of cilengitide as single agent in two phase I trials in adult and pediatric patients [25, 39]. Overall cilengitide was well tolerated, a maximal tolerated dose not reached and a clinical benefit seen both at higher and lower doses. Subsequent trials were initiated with flat (not per m2) twice weekly dosing of cilengitide at an intermediate lower (500 mg) and intermediate higher (2000 mg) dose (Table 1). In a randomized phase II trial of 81 patients with recurrent glioblastoma, treatment was administered at the lower and higher dose. Cilengitide monotherapy was well tolerated independently of the dose. Objective responses were achieved in 5% or 13% of patients treated with 500 mg or 2000 mg, respectively. Progression-free survival after 6 months was 10% and 15%, whereas overall survival was in the range of 6.5 versus 9.9 months, respectively [41]. Recently, long-term survival results were reported demonstrating a 4-year survival rate of 2.4% (95%CI 0.2; 11%) in patients treated with the lower dose, and 10.0% (95%CI 3.2; 21.5%) in patients treated with the higher dose of cilengitide [42]. Similarly, a phase II randomized trial evaluating lower (500 mg) and higher (2000 mg) doses cilengitide added to temozolomide and radiotherapy (TMZ/RT) in newly diagnosed glioblastoma patients was reported in abstract form [43]. Median survival was 18.9 months, with an estimated median survival of 17 months and 21 months, for the lower and higher dose, respectively. Finally, higher exposure of the tumor tissue could be demonstrated in another randomized trial evaluating 3 doses of either 500 mg or 2000 mg cilengitide prior to tumor resection in recurrent disease. Cilengitide concentrations were several-fold higher in the tumor than the corresponding plasma concentrations, demonstrating both excellent penetration through the blood-tumor barrier and retention within the tumor. Importantly, safety could be confirmed and no increase in bleeding complications or wound healing problems were observed [44].

Table 1 Completed trials of cilengitide in primary brain tumors
Full size table

Consistently, the higher dose of cilengitide is favored by exposure, responses, progression-free and overall survival.

Based on a strong preclinical rationale of radiosensitization and synergy with both cilengitide and irradiation and TMZ, we designed in parallel to the above-mentioned trials a phase II pilot trial of adding cilengitide (500 mg) to standard chemoradiotherapy with TMZ (TMZ/RT→TMZ) [45, 46].

In a multicenter pilot study on 52 patients, treatment was well tolerated with no identifiable added toxicity. The primary endpoint of progression-free survival at 6 months compared favorably with historical controls (69% versus 54%) [46, 47]. Median survival was 16.1 months, with a 2-year survival rate of 35% [45]. Interestingly, the outcome was particularly good in patients with a methylated O6-methylguanine–DNA methyltransferase (MGMT) gene promoter, Such patients were previously identified to particularly benefit from TMZ chemotherapy [48]. This is consistent with the working hypothesis, that antiangiogenic therapy may lead to normalization of tumor vasculature, decrease of hypoxia, improved drug delivery and enhanced cytotoxicity. In addition, the currently used higher dose of 2000 mg should also exert relevant direct antitumor activity and directly synergize with standard chemoradiotherapy.

Building on our pilot experience and the consistent encouraging results of the other clinical trials in malignant glioma, Merck KGaA in collaboration with the European Organisation for Research and Treatment of Cancer (EORTC) and the Canadian Brain Tumor Consortium is sponsoring a large prospective clinical phase III trial (CENTRIC study, Fig. 1) [49]. Based on the suggested synergy between normalized perfusion and active chemotherapy, only patients with a methylated MGMT are eligible. This requires centralized upfront MGMT testing, and it is the first prospective international clinical oncology trial incorporating this novel molecular marker to homogenously define the treated patient population. Other changes from the previous phase II trial are the exclusive use of the higher dose of cilengitide (2000 mg), and cilengitide maintenance therapy for up to 18 months. In parallel a separate 3-arm randomized phase II trial has been designed for patients with MGMT unmethylated tumors (CORE study). Here an intensified high-dose daily cilengitide administration before radiotherapy is being explored. Early data indicate that this schedule is well tolerated and safe, and the trial is now ongoing in a multicenter setting. The results of all completed brain tumor clinical trials using cilengitide are summarized in Table 1.

Conclusions

The ongoing phase III CENTRIC trial is successfully recruiting in a multi-national setting and is expected to be completed by the end of the year 2010. Careful analysis of data emerging from this trial will make an important contribution to further defining the role on integrin inhibition in malignant gliomas. The absence of significant toxicity and excellent tolerance will allow to safely testing integrin-inhibition in combination with other targeted and cytotoxic agents. Combined inhibition of integrins and VEGF may lead to further enhancement and increased antitumor effect, and clinical evaluation of this concept is warranted.