evaluates a novel natural product drug compound with significant efficacy in melanoma in vitro and in vivo for preclinical proof of con cept. Methods: A natural product library of 320 plant compounds and extracts was initially screened for anti cancer activity in multi ple cell lines by MTS assay. One lead compound, Withaferin A (WA), was advanced for further analysis. In vitro mechanism of action studies were performed using microarray analysis, Annexin V/PI flow cytometry, and Western analysis for pathway proteins. In vivo analysis was performed on BalbC mice injected s.c. with 5 mil lion B16F10 melanoma cells, allowing tumors to grow and metasta size. Treatment groups included low(2.5 mg/kg/d), and high dose(5 mg/kg/d) treatments i.p. for 3 weeks. Tumor growth and sur vival was followed prospectively. In vitro and in vivo variables were statistically evaluated using SPSS software and significance was defined for p < 0.05. Results: Of all compounds analyzed, WA dem onstrated the highest in vitro cell growth inhibition in melanomas [IC50(B16F10 209.7 6 16 nM); (SKMEL 28 650.1 6 25 nM]]. Microarray and Western analysis confirm modulation of heat shock proteins, shift of cell cycle arrest from G0/G1 arrest to the G2/M checkpoint, induction of apoptosis and down regulation of the MAP Kinase pathway. In vivo studies demonstrate almost complete tumor regression with complete response (CR) in 60% of animals treated with low dose WA (2.5 mg/kg/day) and 80% of animals treated with 5 mg/kg/day WA compared to controls (p < 0.01 for both groups). Additionally at 5/mg/kg dosing, there was complete re versal of organ and lymphatic metastatic disease by inspection after 3 weeks treatment in the 80% of animals completely responding to therapy. Animals demonstrating a complete response had no tumor recurrence or disease progression even 8 weeks after therapy had concluded, suggesting a sustained CR. Conclusion: Withaferin A is a potent anticancer agent with selective activity against mela noma cells in vitro and the ability to reverse metastatic disease in vivo. Future studies will evaluate toxicity and efficacy in other spe cies for preclinical proof of concept as a potentially promising new therapy for advanced melanoma. 16.2. Noninvasive Radiofrequency Field Therapy Induces Cell Death in Human Pancreatic Carcinoma Cells Tar- geted with Gold Nanoparticles. E. S. Glazer, K. L. Massey, S. A. Curley; M.D. Anderson Cancer Center, Houston, TX Introduction. Noninvasive radiofrequency (RF) fields are a type of non ionizing electromagnetic radiation that is known to be safe for human use and to penetrate completely into biologic tissues. Gold or carbon nanoparticles have been shown to release significant heat when exposed to focused shortwave RF fields. We hypothesized that 1) heat release by gold nanoparticles during RF field treatment will be cytotoxic to human cancer cells and 2) the gold nanoparticles can be targeted to specific cancer cells while sparing non targeted cells, thus inducing thermal cytotoxicity only in the targeted cells after RF field treatment. Methods. 20 nm gold particles were conju gated to cetuximab, an epidermal growth factor recpetor 1 (EGFR 1) antibody. A pancreatic carcinoma cell line that highly expresses EGFR 1, Panc 1, and a breast carcinoma cell line that minimally expresses EGFR 1, Cama 1, were treated with 100 nM cetuximab conjugated gold nanoparticles for 3 hours in 60 mm cell culture dishes (n 4). The efficacy of gold nanoparticle uptake in the cells was confirmed by transmission electron microscopy (TEM) and fluo rescence labeling. The cells were placed in a noninvasive RF field with a generator power of 250 W for four minutes. After 36 hours, cell injury and death were evaluated with flow cytometry. Statistical analysis was preformed with Student’s t test. Results. TEM images and flow cytometry confirmed significant intracellular levels of cetuximab targeted gold nanoparticles in Panc 1 cells with almost no gold present in the control Cama 1 cells. Treatment of both cell lines with cetuximab conjugated gold nanoparticles alone (i.e. no RF field therapy) did not alter cell viability (both cell lines >97% viable with or without gold nanoparticle treatment). The EGFR 1 expressing targeted cell line, Panc 1, had a viability of 45.5% 6 11.7% after RF field exposure, while Cama 1 cell had a viability of 91.7% 6 1.6% (See figure, p < 0.008). Importantly, these results were identical when Panc 1 and Cama 1 cells were co cultured in the same plates, treated with cetuximab targeted gold nanopar ticles, and treated in the RF field. Significant gold nanoparticle up take and RF field induced thermal cytotoxicity occurred only in the EGFR 1 positive Panc 1 cells. Conclusions. Using a noninvasive RF field, we show that 1) targeted gold nanoparticles heat and produce thermal cytotoxicity in human pancreatic cancer cells and 2) targeting of gold nanoparticles to spe cific cancer cells leads to RF induced killing of the targeted cells while sparing the non targeted cells. The non targeted cells do not internal ize a sufficient amount of antibody conjugated gold nanoparticles to induce injury in the noninvasive RF field. This interesting finding suggests that this noninvasive RF field therapy may be useful as a minimally toxic cancer treatment if a cancer specific targeting ap proach is utilized to successfully localize gold nanoparticles in only the malignant cells. 16.3. Oncogenic SRC as a Therapeutic Target in Pancreas Cancer. N. S. Nagaraj, H. C. Manning, F. Revetta, M. K. Washington, J. J. Smith, N. B. Merchant; Vanderbilt University, Nashville, TN Introduction: The failure of conventional chemotherapeutic regimes to produce any meaningful impact on survival in patients with pan creatic cancer highlights a desperate need for novel treatment strate gies. The proto oncogene c Src (Src) encodes a non receptor tyrosine kinase whose expression and activity are correlated with advanced malignancy and poor prognosis in a variety of human cancers includ ing pancreas cancer. We have shown that Src expression increase with progression of pancreatic neoplasia. However, the effects of Src inhibition in pancreas cancer remains poorly understood. In the current study, we evaluated the effect of Src inhibition on pancreatic tumorigenesis in vitro and in vivo. Methods: Human pancreatic can cer cell lines BxPC3, PANC1, MiaPaca2, AsPC1, CEPAC, Capan1, Capan2, SW1990 and HPAC were cultured and exposed to varying time points and concentrations of dasatinib (BMS 354825), a novel, orally active, multi targeted kinase inhibitor that targets Src family kinases (SFK). Cell lines were assayed for cell proliferation and ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS ABSTRACTS 195