Adenocarcinoma of the pancreas generally remains an incurable disease by available treatment modalities, demanding the development of a suitable cell-culture/animal model and the discovery and evaluation of novel therapeutic agents. We report the clonal preservation of a human pancreatic cell line (KCI-MOH1) established from a 74-year-old African-American man diagnosed with pancreatic cancer. Initially the human primary tumor was grown as a xenograft in SCID mice and, subsequently, a cell line was established from tumors grown as a xenograft as reported in our earlier publication. The molecular characterization of the primary tumor, the tumors grown as xenograft, and the cell line all revealed similar genotypic properties. By using an automated DNA sequencer, a K-ras mutation (codon 12, GGT to CGT, Gly to Arg) was detected in the pancreatic tumor tissue taken from the patient, whereas no p53 mutation was detected. The same K-ras mutation and unaltered p53 was also found in the xenograft tumor and in the KCI-MOH1 cell line. Chromosome analysis of the cultured cells revealed: 42,XY,add(3)(p11.2),der(7)t(7;12) (p22;q12),-10,-12,add (14)(p11),-18,add (20)(q13),-22/84, idemx2, which is the same chromosome complement found in xenograft tumors. The KCI-MOH1 cell line grows well in tissue culture and forms tumors in the SCID mice when implanted subcutaneously, as well as in orthotopic sites. The KCI-MOH1 cell line-derived SCID mouse xenograft model was used for efficacy evaluation of bryostatin 1, auristatin-PE, spongistatin 1, and gemcitabine alone and in combination. Tumor growth inhibition (T/C expressed as percentage), tumor growth delay (T - C), and log 10 kill for these agents were 38%, 22 days, and 0.53; 15%, 30 days, and 0.80; 24%, 25 days, and 0.66; and 10%, 33 days, and 0.90, respectively. When given in combination, two of seven gemcitabine + auristatin-PE-treated animals were free of tumors for 150 days and were considered cured. Animals treated with a combination of bryostatin 1 and gemcitabine and a combination of spongistatin and gemcitabine produced remissions in only one of seven mice. From these results, we conclude that (a) this is the first study illustrating that clonal characteristics of primary pancreatic tumors remained unchanged when implanted in mice and as a permanent cell line grown in vitro; and (b) there is a synergistic effect between gemcitabine and selected marine products tested in this study, which is more apparent in the gemcitabine and auristatin-PE combination. The results of this preliminary study suggest that these agents should be explored clinically in the treatment of pancreatic cancer.