Groups PBS-control Group (n=6) Ad-VEGF Group (n=5) Ad-VEGF Trap Group (n=5) Adhesion Score (MeanSE) 8.30.5 9.80.5 5.80.6* Vascular Density (lumens/hpf) (MeanSE) 7.61.4 8.531.32 3.5406** *p  0.05 compared to PBS **p  0.05 compared to VEGF, p = 0.076 compared to PBS 30. OPEN LYSIS OF ADHESIONS DECREASES ADHESION REFORMATION MORE EFFECTIVELY THAN LAPARO- SCOPIC LYSIS OF ADHESIONS IN A RAT MODEL. S. G. Prushik 1 , C. B. Aarons 1 , R. Matteotti 1 , K. L. Reed 1 , A. C. Gower 1 , A. F. Stucchi 1 , J. M. Becker 2 ; 1 Boston University School of Med- icine, Boston, MA, 2 Boston University Medical Center, Boston, MA Introduction: Intraabdominal adhesions are a significant and costly complication of abdominal surgery. Up to 94% of patients undergoing open abdominal operations will develop fibrous adhesions. A significant number of these patients will require a second operation for lysis of adhesions (LOA), especially for small bowel obstruction (SBO) where up to 74% of SBO’s requiring surgical intervention are adhesion related. Laparotomy and open adhesiolysis have traditionally been the treat- ment of choice for patients who have failed conservative measures for adhesion related complications. However, minimally invasive proce- dures are increasingly being utilized for laparoscopic exploration and adhesiolysis under the premise that the reduced surgical trauma of a laparoscopic procedure will minimize adhesion reformation. There are no studies to date that have compared adhesion reformation following open vs. laparoscopic LOA. Therefore, the aim of this study was to compare open vs. laparoscopic LOA on adhesion reformation in a rat model that was developed in this lab. We also measured peritoneal fibrinolytic activity following both procedures. Increased peritoneal fi- brinolytic activity due to increased tissue plasminogen activator (tPA) activity and decreased plasminogen activator inhibitor-1 (PAI-1) levels has been correlated with reduced peritoneal adhesion formation. Meth- ods: Intraabdominal adhesions were surgically induced in 28 rats using our ischemic button model. Seven days later, rats underwent laparos- copy (N=16) with carbon dioxide insufflation or laparotomy (N=12) to score and lyse adhesions. Twenty-four hours after LOA, 4 animals from each group (N=8) were sacrificed and peritoneal adhesion tissue and fluid were collected for analysis of tPA and PAI-1 mRNA and total fibrinolytic activity. The remaining 20 animals were sacrificed seven days after LOA and adhesions that had reformed were scored. Results: At the time of the first peritoneal exploration, 78  3.0% of buttons formed adhesions. At the second peritoneal re-exploration, both open and laparoscopic LOA significantly reduced adhesion reformation (p0.05); however, open LOA further decreased adhesion reformation by 60% (p0.05) compared to the laparoscopic LOA group (42  3.2% vs. 17  6.3%). Compared to laparoscopic LOA, total fibrinolytic activity was increased more than 2-fold (2.5  0.64 U/ml vs. 1.2  0.54 U/ml; p = 0.06) in animals undergoing open LOA. Compared with open LOA, tPA mRNA levels were increased by 13% after laparoscopic LOA. Ani- mals undergoing open LOA showed a 50% decrease in PAI-1 mRNA levels (p0.05) and a 2-fold increase in the tPA:PAI-1 ratio (5610.6% vs. 292.7%)(p=0.052) compared to animals undergoing laparoscopic LOA. Conclusions: In this experimental rat model of adhesion formation/reformation, open LOA significantly decreased adhesion ref- ormation compared to laparoscopic LOA. Laparoscopic LOA with car- bon dioxide insufflation appears to compromise the peritoneal fibrino- lytic system and facilitate adhesion reformation. These data suggest that open LOA may be the better choice of operations for the long-term management of adhesion related complications in patients with perito- neal adhesions requiring further surgery. ONCOLOGY I: GENETICS 31. THE ANTIANGIOGENIC HOMEOBOX GENE GAX ACTI- VATES P21WAF1/CIP1 EXPRESSION IN VASCULAR EN- DOTHELIAL CELLS THROUGH INTERACTION WITH UPSTREAM AT-RICH SEQUENCES. Y. Chen, D. H. Gorski; UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ Tumors secrete proangiogenic factors to induce the ingrowth of blood vessels from the surrounding stroma, the end targets of which are vascular endothelial cells (ECs). These proangiogenic factors bind to cell EC surface receptors and activate intracellular signaling pathways, the end targets of which are transcription factors, which in turn activate and/or repress downstream batteries of genes in order to induce an activated, angiogenic phenotype. Gax, a homeobox gene, has been shown to inhibit angiogenesis and induce p21WAF1/ CIP1 expression in vascular ECs. In order to elucidate further the mechanism through which Gax activates p21WAF1/CIP1 expression, we constructed Gax cDNAs with deletions of either the N-terminal domain, the homeodomain, or the C-terminal domain, and then assessed these constructs for their ability to activate p21WAF1/ CIP1. We found that there was an absolute requirement for the homeodomain. Deleting the C-terminal domain decreased, but did not abolish, transactivation of the p21WAF1/CIP1 promoter, while deleting the N-terminal domain or the opa repeat in the N-terminal domain did abolish transactivation. Next, we performed chromatin immunoprecipitation and found approximately 15 kb upstream of the p21WAF1/CIP1 ATG an ATTA-containing Gax binding site (des- ignated A6) with a sequence similar to that of other homeodomain binding sites. Gax was able to bind to A6 in a homeodomain- dependent manner and thereby activate the expression of a reporter gene coupled to this sequence, and this activation was abolished by mutating specific residues in this sequence. We conclude that Gax activates p21WAF1/CIP1 at least in part through an upstream AT- rich sequence that appears to act as an enhancer. Given the multiple biological activities of Gax in regulating EC function, identification of a putative Gax binding site will allow study of how Gax activates or represses other downstream targets to inhibit tumor-induced angio- genesis. Gax or its downstream targets may represent a target for the antiangiogenic therapy of cancer. 32. EFFECTIVE ABLATION OF HUMAN PANCREATIC CAN- CER CELLS IN SCID MICE USING SYSTEMIC ADENO- VIRAL RAT INSULIN PROMOTER DIRECTED THYMI- DINE KINASE GENE THERAPY. S. Liu 1 , Z. Li 1 , A. R. Davis 2 , E. A. Davis 2 , X. Wang 1 , F. C. Brunicardi 1 ; 1 Michael E. Debakey Department of Surgery, Baylor College of Medicine, Houston, TX, 2 PEDI/HEM-ONC Cell & Gene, Baylor College of Medicine, Houston, TX Introduction: Our previous study has shown that adenoviral- mediated rat insulin promoter-thymidine kinase gene therapy (Ad- RIP-TK) and ganciclovir (GCV) were cytotoxic for human pancreatic cancer cells in vitro. The purpose of this study was to determine whether systemic Ad-RIP-TK and GCV would effectively ablate hu- man pancreatic cancer cells in SCID mice. Methods: SCID mice were inoculated with 10 6 MIA PaCa 2 cells/mouse via intraperitoneal injection (ip). Two weeks later, the mice received either 1) 2 wks GCV only (n=12) , 2) Ad-RIP-TK only (n=12), 3) iv 10 7 viral particles Ad-RIP-TK followed by 2 wks of GCV (10 7 Ad-RIP-TK/2wksGCV), 4) iv 10 8 vp Ad-RIP-TK followed by 2 wks GCV(10 8 Ad-RIP-TK/ 2wksGCV) or 5) iv 10 8 vp Ad-RIP-TK followed by 4 wks GCV(10 8 Ad-RIP-TK/4wksGCV). Ad-RIP-TK was injected via tail vein and GCV was given at a dose of 40mg/kg ip bid. Data were collected at 2, 4, 7, 14, 30, 60, 90, and 120 days after treatment. Tumor volume was evaluated at necropsy and statistically analyzed using student T test (P0.05). Serum glucose and insulin levels were examined and tis- sues were saved for pathological analysis. Survival analysis was 163 ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS