Moving Beyond Anti–Vascular Endothelial Growth Factor Therapy in Ovarian Cancer

Ovarian cancer remains associated with high rates of morbidity and mortality for many reasons, including advanced stage at diagnosis and inherent or emergence of drug resistance. Over the past several decades, the primary treatment paradigm has evolved to incorporate aggressive primary surgical cytoreduction, followed by paclitaxel and platinum-based chemotherapy. Promising variants of the latter include intraperitoneal and weekly chemotherapy regimens. However, despite progressive gains in median survivorship, cure rates, as a cohort, have not measurably improved, even in the setting of chemotherapy-sensitive tumors. Thus, the vast majority of women with advanced ovarian carcinoma will experience relapse and eventually develop drug-resistant disease. Better therapeutic agents and strategies are urgently needed. An attractive target for therapeutic intervention has been the tumor microenvironment, particularly the tumor vasculature, as it has been hypothesized that endothelial cells are genetically more stable than tumor cells. In addition, tumor neovascularization is a highly regulated process that incorporates a number of proangiogenic molecular targets, such as vascular endothelial growth factor (VEGF). While the influence of these growth factors drives complex development of neovascularization (both sprouting and nonsprouting angiogenesis) to the growing tumor mass, the expanding availability of drug targets within the tumor microenvironment has provided an opportunity to further refine the options for effectively controlling tumor growth. In patients with ovarian cancer, the best-characterized therapeutic strategy in this regard has been the VEGF/VEGF-receptor axis. On the basis of promising preclinical studies targeting VEGF, clinical research was launched in patients with recurrent disease. The first agent to enter the clinical domain that demonstrated clear phase II, single-agent activity was bevacizumab, a monoclonal antibody targeting VEGF. Measures of success in these trials were objective response rates (ranging from 16% to 21%) and delay in tumor progression (up to 41% at 6 months). These encouraging phase II studies provided strong rationale for phase III investigations, which were launched in both the primary and recurrent setting. The first of these phase III studies to mature was Gynecologic Oncology Group (GOG) -218 (three-arm trial: paclitaxel/carboplatin chemotherapy (CT) versus CT plus concurrent bevacizumab versus CT plus concurrent and maintenance bevacizumab). A similar question is being addressed by Gynecologic Cancer Intergroup (GCIG) International Collaborative Ovarian Neoplasm 7 trial (ICON7; two-arm trial: CT with or without concurrent and maintenance bevacizumab). Both of these trials were conducted in the front-line adjuvant setting after tumor debulking. Despite the many differences between the two trials, including trial design, patient eligibility, use of a placebo control, and dose and duration of bevacizumab, they both produced modest but statistically significant gains in progression-free survival (PFS). Analysis of overall survival (OS) for both studies is immature but, at least for GOG 218, is not expected to be measurably different among the treatment arms. In the recurrent setting, a promising impact on PFS was recently reported with bevacizumab when combined with gemcitabine and carboplatin in patients with platinum-sensitive disease. An important design difference in this trial (OCEANS) compared to GOG-218 and ICON7 was the opportunity to administer bevacizumab maintenance until disease progression. While it is not yet known if this strategy will affect OS, the modest benefits observed in these, as yet unpublished, studies coupled with the high cost of bevacizumab therapy leave room for alternative antiangiogenic approaches. In addition, our growing understanding of antiangiogenic escape mechanisms suggests that additional pathways involved in tumor neovascularization will need to be targeted in order to significantly improve clinical efficacy. Alternative approaches for blocking tumor neovascularization would ideally address not only parallel mechanisms, but also those that are induced after exposure to anti-VEGF/VEGFR drugs. Many such targets have been discovered, including platelet-derived growth factor, fibroblast growth factor, focal adhesion kinase, the polycomb group protein enhancer of Zeste homolog 2, notch family, and the angiopoietin (Ang) -TIE system. The Ang-TIE pathway has been of particular interest as a result of its roles in angiogenesis and inflammatory pathways. It consists of two type I tyrosine kinase receptors (TIE1 and TIE2) and three secreted ligands in humans (Ang1, Ang2, and Ang4). Ang1 functions as a TIE2 receptor agonist, promoting vascular normalization and decreasing new blood vessel formation. In contrast, Ang2 normally functions as an Ang1 antagonist, resulting in increased angiogenesis. Ang1 is constitutively expressed in many adult tissues, whereas Ang2 is predominantly expressed in tissues undergoing remodeling. In ovarian cancer, interplay between VEGF and Ang2 has been shown to play a direct role in supporting tumor angiogenesis. Some studies have suggested that high Ang2 levels may limit the efficacy of anti-VEGF treatment. A number of strategies for therapeutically blocking the Ang-TIE system have been developed, JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L S VOLUME 30 NUMBER 4 FEBRUARY 1 2012