New molecular targets and novel anticancer treatments: emerging trends in neuro-oncology

As with other cancers, gliomas arise as a result of an accumulation of multiple genetic aberrations in normal precursor cells. These alterations have been determined to occur through chromosome loss, gene amplification, gene rearrangements and point mutations. Several of these genetic changes have been shown to promote the aggressive phenotypic characteristics associated with high-grade tumor malignancy.

The treatment of these neoplasms needs a multimodal approach; surgery and radiation have had the greatest impact on increasing survival. Recently, even chemotherapy has been shown to increase survival in patients with newly diagnosed glioblastoma [1]. Unfortunately, this kind of tumor is still incurable, with a short time to progression and neurological deterioration. Currently, the aim of treatment is to improve neurological deficits and to increase survival, while maintaining the best possible quality of life.

For these reasons, new drugs with better activity/toxicity profiles have been synthesized, based on recent biological acquisitions in molecular oncology and pharmacology.

In recent years, the field of cancer therapy has witnessed the emergence of novel targeted strategies that inhibit specific cancer pathways and key molecules in tumor growth and progression. Several putative biological factors have been studied as candidate prognostic factors in neuro-oncology, contributing information to the plethora of biological knowledge in the oncological community.

There are at least two major reasons to consider the neuro-oncological setting as a peculiar model for testing these new agents. First, this branch of clinical oncology remained in a dormant status for many years owing to the lack of effective cytotoxic treatments. However, in the last few years we have observed a rapid parallel exponential evolution of clinical treatments (e.g., new alkylating agents and targeted therapies), brain-imaging technology and translational research, and now neuro-oncology can be considered a model without the heavy bias of old Phase III trials that did not consider the molecular behavior of tumors.

In addition, classical end points to assess activity in clinical oncology do not seem to be appropriate to evaluate the novel, molecularly targeted agents with a prevalent cytostatic activity. Neuro-oncology represents a particular model, even in this setting. In fact, the peculiar characteristics of gliomas, with the frequent presence of necrosis and the use of corticosteroids that can alter imaging characteristics, represent a brain-imaging dilemma. The recognition of this problem has led to the development of alternative end points (such as 6-month, progression-free survival rate [PFS-6]) and response criteria (Macdonald’s criteria [2]) that took into account clinical conditions and corticosteroid use in clinical trials evaluating treatment of recurrent glioblastoma. Therefore, the absence of progression, rather than the reduction in tumor, is the criterion for establishing efficacy. A series of eight Phase II studies were compiled and a PFS-6 of 15% or less was set as the cut-off point to define an ineffective regimen for recurrent glioblastoma [3]. Later studies used this cornerstone in devising a statistical design that incorporated both the definition of efficacy and an adequately powered analysis for determining the optimal sample size for a clinical trial. This approach may be particularly useful, even for evaluating treatments such as signal-transduction pathway modulators, which may be cytostatic (halting tumor growth) rather than cytotoxic (causing tumor shrinkage). For this reason, neuro-oncology offers a model with historical comparators with new agents’ trials.

However, the treatment of brain tumors poses several unique challenges. The blood–brain barrier is variably functional, with heterogeneity demonstrated by brain-imaging studies, such as magnetic resonance imaging. Also, concurrent medications, particularly anticonvulsants, can profoundly impact the pharmacokinetics of several chemotherapeutic agents. A number of anticonvulsants (i.e., phenytoin, carbamazepine and phenobarbital) induce hepatic cytochrome P450 enzymes, markedly altering chemotherapy metabolism and clearance (e.g., irinotecan [4]).

Based on this background, several new agents that target specific signal-transduction pathways or molecular targets have been evaluated in neuro-oncology.

As in lung cancer, the epidermal growth factor receptor (EGFR) is an attractive therapeutic target, even in glioblastoma multiforme. The EGFR gene is amplified in 40% of primary glioblastomas [5], frequently resulting in overexpression of the EGFR protein, often with its vIII mutation. In general, the studies with EGFR inhibitors show that these agents were well tolerated. However, the responses tend to be limited and short lived. Moreover, it has not been possible to correlate response to a specific molecular phenotype, partly as a result of the low number of responders. At least three reports have been published recently on the use of EGFR tyrosine kinase inhibitors (gefitinib and erlotinib). However, these studies were conducted on small samples of patients and sometimes on retrospective data and showed heterogeneous response rates and heterogeneous putative molecular predictors of EGFR-inhibitor activity [6–8].

Globally, this first generation of studies on small-molecule inhibitors of EGFR tyrosine kinase did not provide the expected results in patients with high-grade gliomas. As with other targeted agents, it is likely that EGFR inhibitors may be more effective when combined with other therapies, such as radiation therapy or other targeted molecular agents.

Platelet-derived growth factor receptor (PDGFR) tyrosine kinase inhibitors have also been tested in brain tumor clinical trials. PDGFR overexpression has been described in up to two-thirds of glioblastomas [9] and STI-571 (imatinib mesylate; Gleevec™) has shown activity against glioblastoma cell lines in preclinical in vitro studies. The European Organization for the Research and Treatment of Cancer (EORTC) has conducted separate trials for patients with anaplastic astrocytoma and glioblastoma, with a PFS-6 of 15% [10,11]. Superior efficacy was suggested for the combination of imatinib and hydroxyurea in an uncontrolled trial [12]and a Phase II study [13].

Other agents against different intracellular targets were evaluated and are under current evaluation in many trials, such as the mammalian target of rapamycin (mTOR) inhibitor temsirolimus, farnesyltransferase inhibitor tipifarnib and multitarget inhibitors, such as sorafenib (RAF, vascular endothelial growth factor receptors [VEGFRs] and PDGFR tyrosine kinase inhibitor). Recent approaches to cancer treatment have been based on blocking angiogenesis, which is required for tumor survival and growth. Marked neovascularization is a key feature of glioblastoma, which is characterized by an exuberant pattern of vascular endothelial cell proliferation and tumor necrosis [14]. Increased vascularization correlates with the invasive properties of gliomas and, ultimately, with a more malignant tumor phenotype. Angiogenesis is a complicated, multistep process regulated by the balance of endogenous stimulatory and inhibitory factors elaborated by tumor cells and by other cell types. VEGF is the most potent angiogenic molecule, thus the VEGF pathway is considered a primary target for antiangiogenic compounds. Blockage of the VEGF pathway has been clearly shown to have antitumor activity in animal models. In rats, vatalanib (PTK787/ZK222584), an oral VEGFR tyrosine kinase inhibitor, was shown to slow glioma development [15]. A clinical trial investigating the addition of PTK787/ZK222584 to standard temozolomide and radiotherapy in newly diagnosed glioblastoma patients is ongoing by the EORTC. Recently, the anti-VEGF monoclonal antibody bevacizumab, in combination with irinotecan, has shown promising activity in a Phase II trial in 32 malignant glioma patients, with a PFS-6 of 38% (30% in glioblastoma patients) [16].

As we have shown, even in the neuro-oncology setting, there is now an embarrassment of riches [17] and the next few years will tell us if all these therapeutic weapons are sufficient to enter a ‘Golden Age’ in the treatment of these fatal diseases.