Abstracts
2001 Digestive Disease Week

# 1788 Human Ductal Pancreatic Adenocarcinoma Cells Can Be Targeted And Destroyed By A Rat Insulin Promoter Thymidine Kinase Construct, In Vivo
Thomas A. Tirone, Xaioping Wang, Nancy S. Templetone, F Charles Brunicardi, Houston, TX

Purpose: Preliminary data previously presented suggests that PDA cells can be selectively targeted and killed using the rat insulin promoter (RIP) driving the suicide gene thymidine kinase (tk) in vitro, and that the presence of the transcription factor PDX-1 in these PDA cells is vital for insulin promoter activation. The purpose of this study was to investigate if human PDA cells could be targeted and killed in a mouse PDA tumor model and to determine if PDX-1 is present in PDA cells freshly resected from humans.

Methods: 5 x 10⁵ PANC-1 cells were injected intraperitoneally (IP) into ICR/scid mice. RIPtk and the reporter gene RIPlacZ were generated. For in vivo gene delivery the DNA was complexed with 20mM extruded DOTAP:cholesterol and delivered IP. Twenty-one days post tumor injection the mice were randomized to receive either RIPtk, RIPlacZ, or nothing. All mice received 7 days of ganciclovir (GCV) 40mg/kg IP BID. A subset of mice with tumors that received RIPlacZ were sacrificed at 72 hours and their tissues were fixed and stained with X-gal. Three human PDA tumors were obtained; two hepatic metastases and one peripancreatic nodule, and western blots were performed with an antibody to PDX-1.

Results: PDA cells were targeted in vivo with RIPlacZ; only the tumor cells stained blue. Mice that received RIPtk had a significant reduction in tumor burden; the combination eliminated all microscopic disease in 8/9 mice (See table). All three fresh human PDA specimens contained the transcription factor PDX-1.

Conclusion: The data suggest that RIP can be used to target and kill PDA cells in vivo. The presence of PDX-1 in freshly resected human PDA tumors further suggests that this therapy may possibly be effective in the treatment of human pancreatic ductal adenocarcinoma in the future.

Tumor Size

Vector Tumor (Gross) Tumor (Micro) Size*

RIPtk 0/9 1/9 <1mm^L

RIPlacZ 9/9 9/9 15x20x30mm

None 9/9 9/9 20x22x45mm

p<0.05 Chi Square

^Dp<0.05 ANOVA

*Tumor size based on the average of the largest tumor for each of the nine mice in each group