Previously, a Bcl-2-interacting protein, BAG-1, was cloned from mouse cells and was shown to interact with several other proteins and to be important for inhibition of apoptosis. Human BAG-1 (hBAG-1) cDNA, recently isolated by us and two other groups, has been shown to be identical to a hormone receptor-binding protein, RAP46. However, di erent molecular masses of hBAG-1 protein products were noted by these three groups. Here we demonstrated that hBAG-1 protein was expressed as four isoforms, designated p50, p46, p33 and p29, with apparent molecular masses of 50 kDa, 46 kDa, 33 kDa and 29 kDa, respectively. Deletion, site-directed mutagenesis and in vitro transcription/translation analysis showed that the four protein products of hBAG-1 were expressed by alternative initiation from four di erent start codons through a leaky scanning mechanism. Furthermore, we demonstrated that the distinct forms of hBAG-1 have di erent subcellular localizations, suggesting that they may have distinct functions in the cells. Characterization of hBAG-1 RNA and protein also showed that hBAG-1 was overexpressed in human cervical, breast and lung cancer cell lines. Taken together, these data clarify the con¯icting observations reported in the literature and suggest that hBAG-1 is expressed as four forms of protein products, which may play a di erential role in apoptosis and oncogenesis of human cells.

ANG1005, a novel taxane derivative, consists of 3 paclitaxel molecules covalently linked to Angiopep-2, designed to cross the blood-brain and bloodcerebrospinal barriers and to penetrate malignant cells via LRP1 transport system. Preclinical and clinical evidence of efficacy with ANG1005 has been previously shown. EXPERIMENTAL DESIGN A multi-center, open-label phase 2 study in adult patients with measurable recurrent brain metastases from breast cancer (BCBM), with or without leptomeningeal carcinomatosis (LC) was conducted (n=72 BCBM; n=28 LC subset). ANG1005 was administered intravenously (IV) at 600 mg/m 2 q3w. Tumor assessment was based on CNS RECIST 1.1 for intracranial, and RECIST 1.1 for extracranial response. The primary endpoint was determination of intracranial objective response rate (iORR). RESULTS Median age was 47.5 years. Safety profile was similar to that of paclitaxel with myelosuppression as the predominating toxicity. Average number of prior CNS directed therapies was 2.8 and 94% of the patients had prior taxane treatment. Patient benefit (stable disease or better) was seen in 77% (intracranial) and 86% (extracranial) of the evaluable patients, with iORR of 15% (investigator) or 8% (independent radiology review). In the LC subset, 79% of the patients had intracranial disease control and estimated median overall survival of 8.0 months (95% CI 5.4-9.4). CONCLUSION Even though the study pre-set rule for iORR per IRF was not met in this heavily pretreated population, a notable CNS and systemic treatment effect was seen in Research.

Anti-angiogenesis is currently considered as one of the major antitumor strategies for its protective effects against tumor emergency and later progression. The anti-diabetic drug metformin has been demonstrated to significantly inhibit tumor angiogenesis based on recent studies. However, the mechanism underlying this anti-angiogenic effect still remains an enigma. In this study, we investigated metformin-induced inhibitory effect on tumor angiogenesis in vitro and in vivo. Metformin pretreatment significantly suppressed tumor paracrine signaling-induced angiogenic promotion even in the presence of heregulin (HRG)-β1 (a co-activator of HER2) pretreatment of HER2+ tumor cells. Similar to that of AG825, a specific inhibitor of HER2 phosphorylation, metformin treatment decreased both total and phosphorylation (Tyr 1221/1222) levels of HER2 protein and significantly reduced microvessel density and the amount of Fitc-conjugated Dextran leaking outside the vessel. Furthermore, our results of VEGF-neutralizing and -rescuing tests showed that metformin markedly abrogated HER2 signaling-induced tumor angiogenesis by inhibiting VEGF secretion. Inhibition of HIF-1α signaling by using RNAi or YC-1, a specific inhibitor of HIF-1α synthesis, both completely diminished mRNA level of VEGF and greatly inhibited endothelial cell proliferation promoted by HER2+ tumor cell-conditioned medium in both the absence and presence of HRG-β1 pretreatment. Importantly, metformin treatment decreased the number of HIF-1α nucleus positive cells in 4T1 tumors, accompanied by decreased microvessel density. Our data thus provides novel insight into the mechanism underlying the metformin-induced inhibition of tumor angiogenesis and indicates possibilities of HIF-1α-VEGF signaling axis in mediating HER2-induced tumor angiogenesis.