An orthotopic model of human pancreatic cancer in severe combined immunodeficient mice: potential application for preclinical studies.

Pancreatic adenocarcinoma is one of the most incurable and least understood of all human cancers. It is the fourth leading cause of cancer-related mortality in males (after lung, prostate, and colon) and in females (after lung, breast, and colon) in the United States with <2-3% of patients surviving >5 years. In an attempt to search for more effective therapies for this disease, we report here, for the first time, an effective treatment, the combination of gemcitabine and auristatin-phenethylamine (PE), against an orthotopic implantation of a human pancreatic adenocarcinoma cell line (HPAC) in severe combined immunodeficient (SCID) mice. Tumor implantation was performed by injecting 100 microl of the HPAC cell suspension (1 x 10(6) cells) directly into the pancreas of 5-week-old SCID mice. After implantation, tumor formation was checked twice a week. All palpable tumors were detected within 21 days (100% take rate), and tumors were confirmed histologically to be pancreatic adenocarcinoma. For the subsequent efficacy trial, tumor-bearing SCID mice were randomized into four groups with five mice in each group. One served as a control, the second received gemcitabine alone (2.5 mg/kg/injection i.p.), the third received auristatin-PE alone (2.0 mg/kg/injection i.v.), and the fourth group received the combination of gemcitabine (i.p.) and auristatin-PE (1.5 mg/kg/injection i.v.). All animals were euthanized 7 days after the completion of their treatments, and the pancreases were resected. Histological examination revealed the tumors to be adenocarcinoma. The tumors were composed of diffuse sheets of cells interrupted by glandular spaces containing secretory material. Cytologically, the tumor cells were large, pleomorphic, and hyperchromatic. Many cells contained intracellular lumina containing mucin. Immunohistochemical studies showed strong p21WAF1 (p21) expression but no immunoreactivity with p53 and Her-2/neu antibodies. The mean pancreatic weight in the gemcitabine/auristatin-PE combination group was significantly (P = 0.014) lower (0.84 +/- 0.639 g) when compared with those of the control (2.91 +/- 1.19 g) and gemcitabine alone (1.84 +/- 0.796 g; P = 0.064) groups. In addition, the mean weight in the combination group approached statistical significance when compared with the auristatin-PE group alone (1.16 +/- 0.635 g; P = 0.028). We conclude that the combination of gemcitabine and auristatin-PE is an effective treatment against HPAC tumors in this xenograft model and more effective than treatment with either gemcitabine or auristatin-PE alone and could be considered for future animal studies with pancreas cancer and/or for human clinical trials.