Preclinical evaluation of huC242-DM4 in tumor xenograft models of canag-positive human gastric cancer

B53 HuC242-DM4 is a novel, targeted anticancer agent in development for the treatment of CanAg-expressing tumors. It is formed by the conjugation of the potent maytansinoid cell-killing agent, DM4, to the CanAg-binding humanized antibody, huC242. HuC242-DM4 currently is in Phase II testing for the treatment of gastric and gastroesophageal junction cancer. The rationale for the clinical trial is provided by in vitro characterization including immunohistochemistry studies, and in vivo preclinical evaluation of the compound. To identify the distribution of the CanAg on human esophageal and gastric cancer, immunoreactivity of the murine C242 antibody was assessed with formalin-fixed, paraffin-embedded human tumor samples. Nine out of eleven (82%) patients with adenocarinoma and eight out of nine (89%) patients with squamous cell subtype stained positive with murine C242 antibody on esophageal tumor specimens. Thirteen of twenty-one (62%) gastric patient specimens stained positive with the antibody. Our preclinical assessment of its activity included evaluation in xenograft tumor models. In mice bearing NCI-N87 (human gastric carcinoma, CanAg+) tumors, huC242-DM4 caused complete tumor regressions with single intravenous administration as low as 3.5 mg of conjugate protein per kg, well below the estimated maximal tolerated dose of about 50 mg of conjugate protein per kg in mice. The lowest single dose tested, 1.8 mg/kg, resulted in log 10 cell kill of 1.27. At the highest dose tested, a single injection of 18.9 mg/kg, huC242-DM4 caused complete tumor regressions in 5 of 6 mice with a mean tumor-free period of 55 + 36 days. No body weight loss was observed at any dose. Similar results were found in a second model of gastric carcinoma using xenografts of the SNU 16 cell line. For active doses, the pharmacokinetics and the area under the curve (AUC) of tumor drug exposure in mice were compared to the pharmacokinetics and AUC of huC242-DM4 in Phase I human testing. The analysis indicates that the Phase II dose of 168 mg/m 2 in humans is likely to achieve the blood levels (plasma exposure) found to be efficacious in preclinical evaluation.