The Adnectin CT-322 is a novel VEGF receptor 2 inhibitor that decreases tumor burden in an orthotopic mouse model of pancreatic cancer

Pancreatic cancer continues to carry a poor prognosis, with a 5-year survival rate of approximately five percent [1]. As patients typically present at an advanced stage, new chemotherapeutic strategies are necessary to improve the dismal prognosis associated with this disease. Angiogenesis is a hallmark of cancer [2] and is required for cancer growth beyond 1–2 mm³ [2, 3]. Vascular endothelial growth factor A (VEGF) is the best characterized member of the VEGF family of growth factors. VEGF is a potent angiogenic factor expressed during development and in tumors [4, 5]. The effects of VEGF are mediated by binding to one of its two receptors VEGF receptor 1 or 2 (VEGFR1, VEGFR2) [4, 6]. Tumor angiogenesis is driven primarily by VEGF:VEGFR2 interaction [5, 6]. The effect of VEGFR1 activation is less understood, but is thought to be involved in macrophage chemotaxis [5‐7].

The complexity of the VEGF pathway allows for multiple targets for inhibiting tumor angiogenesis [5, 8]. For example, bevacizumab (Avastin^®, Genentech, Inc., South San Francisco, CA) is a monoclonal antibody to human VEGF which binds VEGF and blocks its interaction with both VEGFR1 and VEGFR2 [9]. Bevacizumab has been shown to be effective in combination with chemotherapy for the treatment of metastatic colorectal cancer and non-small cell lung cancer [10, 11]. Receptor tyrosine kinase inhibitors have also been developed which inhibit the VEGF receptors [5, 8]. These small molecules penetrate into cells and, unlike antibodies, inhibit multiple members of the VEGF receptor family. This broad spectrum of inhibition may lead to different side effect profiles from monoclonal antibodies [5].

There are a variety of proteins being developed as new biologic drugs beyond the traditional biologic class of monoclonal antibodies [12]. Adnectins are a new class of targeted biologics among the most advanced of such proteins. Adnectins are well-suited to pharmaceutical discovery and development, based on preclinical data [13, 14]. These small proteins are derived from the 10th type III domain of human fibronectin, an extracellular protein that is abundant in human serum and the extracellular matrix, and naturally binds to other proteins [13, 14]. By changing the amino acid sequence of three targeting loops clustered at one end of the protein, an Adnectin can be designed to bind to a specific disease target, such as a receptor, ligand or protein with nanomolar or picomolar affinity, and potency and specificity comparable to or better than antibodies. One such Adnectin has been developed that binds to VEGFR2. This construct, CT-322, has been shown previously to block the activity of murine and human VEGFR2 in vitro [13].

In the present study, we were interested in whether this novel compound would block tumor angiogenesis and subsequent growth in an orthotopic model of pancreatic cancer. In the following experiments, we demonstrate that CT-322 is effective at treating pancreatic tumors in two animal models, that CT-322 blocks tumor angiogenesis and that treatment with CT-322 induces tumor destruction.

Pancreatic cancer remains a deadly disease, with a dismal five year survival [1]. Contributing to the poor outcome is the resistance or insensitivity of most pancreatic cancers to chemotherapy. Gemcitabine, the current standard of care, extends life only minimally. Therefore, novel therapeutics agents are necessary to improve the survival associated with this disease.

Angiogenesis has been described as both a hallmark of cancer [2] and as an "organizing principle" around which anti-cancer therapy can be developed [20]. VEGF is a key mediator of tumor angiogenesis that functions primarily through VEGFR2. The complexity of the VEGF signaling pathway allows for multiple levels of inhibition, as seen by the large number of lead compounds entering trials that target this pathway [8, 20]. One approach for VEGF inhibition is to reduce the availability of ligand (VEGF), as demonstrated by bevacizumab, a monoclonal antibody specific for human VEGF [9]. This approach has been validated in clinical trials combining chemotherapy and bevacizumab in the treatment of metastatic colorectal [10] or breast cancer [21]. However, despite initially promising results in pancreatic cancer, a subsequent randomized controlled trial failed to show clinical benefit of combining bevacizumab with gemcitabine [22, 23].

The major finding of this study is that the adnectin CT-322 effectively controls the growth of orthotopic pancreatic tumors in mice by reducing VEGF-induced angiogenesis. We evaluated the efficacy of CT-322 in xenograft (MiaPaCa-2) and syngeneic (Pan02) tumors. As a single agent CT-322 appears to be more effective in the xenograft tumors where it reduced tumor growth by 50% whereas growth was reduced by less than 25% in the syngeneic model and was not different from tumors from control-treated animals. It is worth pointing out that Pan02 tumors are more invasive locally than MiaPaCa-2 tumors and often animals with Pan02 tumors need to be sacrificed due to disease burden that is far smaller than what is seen in xenograft models. For instance MiaPaCa-2 tumors from control-treated animals were on average over 1 g in size whereas Pan02 tumors from control-treated animals were on average just over 0.4 g in size. Thus it is possible that the xenograft tumors are more dependent upon VEGF-induced angiogenesis than the Pan02 tumors and thus anti-VEGF strategies are more effective as standalone agents in this setting.

We also observed in both tumor models that CT-322 as a single agent reduced microvessel density. Interestingly, Pan02 tumors from mice treated with CT-322 and gemcitabine did not show the same degree of reduction in microvessel density as tumors from mice treated with CT-322 alone. This is perplexing considering that the combination therapy was more effective at reducing overall tumor size. One possible explanation is that not all the blood vessels stained with MECA-32 are functional. In future studies this could determined by intravenous injection of an endothelial cell marker (eg., tomato lectin) or a fluorescent dye to track perfusion. An alternative non-exclusive explanation is that a high percentage of the blood vessels present in the tumors in the combination treated group have been co-opted from normal pancreas and are associated with pericytes and thus less sensitive to anti-VEGF therapy. This is plausible given that the size of the tumors is quite small in the combination treated group. It is worth pointing out that the tumor weights shown in figures 3 and 4 include residual normal pancreas, which typically weighs 0.1–0.15 g. It is difficult especially in the Pan02 model to dissect the tumor from the normal tissue due to the local invasion by the tumor. Thus, especially in the syngeneic model we may be underestimating the efficacy of CT-322-based therapy.

CT-322 has begun clinical trials and initial results from the first phase I studies were discussed recently at the 2008 American Society of Clinical Oncology meeting [24]. Thirty nine patients with solid tumors or non-Hodgkin's lymphoma were enrolled. 2 mg/kg/week was determined to be the maximum tolerated dose and out of 37 patients evaluated 49% had stable disease. These are promising results and have prompted a phase II study in combination with irinotecan in patients with recurrent glioblastome multiforme [24].

Our results show that in two aggressive preclinical pancreatic cancer models, the strategy of selectively targeting VEGFR2 with CT-322 is effective. This compound is experimentally useful in that it blocks both mouse and human VEGFR2, thus allowing us to demonstrate the effectiveness of this drug in a fully immunocompetent mouse model. We also show that CT-322 enhances the efficacy of gemcitabine in this setting. As gemcitabine typically only prolongs life by a few months, agents improving its efficacy are needed. Further, these data demonstrate the safety and therapeutic utility of the Adnectin platform and suggest that further investigation of CT-322 is warranted in both animal models and clinical trials.