Vascular disrupting agent STA-9584 selectively targets microvasculature at both the center and periphery of tumors

1444 Background: Vascular disrupting agents (VDAs) preferentially block blood flow in tumors relative to normal tissues, thereby inducing tumor hypoxia and necrosis. This is distinct from antiangiogenic therapies, which inhibit new blood vessel growth and promote vascular normalization. However, first-generation VDAs display poor single-agent efficacy in preclinical models due to the persistence of cancer cells at the well-perfused “viable rim” of tumors, which allows rapid tumor regrowth to occur. STA-9584 is a novel VDA with increased potency and an improved therapeutic index in preclinical models relative to VDAs such as combretastatin A4 phosphate (CA4P). To examine the in vivo mechanisms of action of STA-9584 and CA4P, we directly compared their activities in mouse tumor models. Results: In the mouse EMT6 syngeneic breast cancer and human PC-3 prostate cancer models, STA-9584 demonstrated superior inhibition of tumor growth relative to CA4P at each compound’s maximum tolerated dose (MTD). In the highly aggressive EMT6 model, %T/C values for STA-9584 and CA4P were 3 and 55, respectively, when measured 5 days after single drug doses. Histological analysis of EMT6 tumors showed that STA-9584 elicited a 73% increase in total necrotic area when compared with CA4P at 24 hours after drug treatment. In the viable rim, this was accompanied by a 77% decrease in CD31 + microvasculature cells and a 7-fold increase in TUNEL + apoptotic tumor cells with STA-9584 relative to CA4P. To assess the effects of these agents on tumor blood flow, a dye extravasation assay was performed on EMT6 tumors, showing that STA-9584 and CA4P inhibited blood flow by 61% and 43% at 4 hours after drug treatment, respectively. Similar results were observed in the PC-3 model by imaging of a near infrared vascular probe using fluorescence molecular tomography. Since tumors display regional differences in the extent of vascular perfusion, EMT6 tumors were also imaged with a microbubble contrast agent using high-resolution micro-ultrasound. Treatment with STA-9584 was found to completely block blood flow in the highly perfused subregions of tumors. To demonstrate that vascular disruption by STA-9584 was selective for tumor microvasculature, in vitro and in vivo toxicity studies were conducted in mice, rats and dogs. STA-9584 was found to have low in vitro cytotoxicity towards normal cells and an acceptable in vivo safety profile. The plasma exposure level of STA-9584 at the MTD exceeded the minimum efficacious dose exposure level, and cardiovascular toxicity was not dose-limiting. Conclusions: STA-9584 is a promising new VDA that blocks blood flow by specifically disrupting tumor microvasculature, not only in the center, but also at the periphery of tumors. Unlike CA4P, STA-9584 has a positive therapeutic index in preclinical models, and cardiovascular toxicity is predicted not to be dose-limiting.