Objectives: Negative feedback regulation of pancreatic proteases controls pancreatic secretion in most species and pancreatic growth in rodents. Its mechanism involves the inhibition of intraluminal proteases, resulting in sustained elevation of plasma cholecystokinin (CCK) concentrations, producing a chronic trophic stimulus to the pancreas that leads to the formation of pancreatic nodules and adenomas. Ximelagatran, whose active form, melagatran, inhibits both thrombin and the serine protease trypsin, is under clinical development as an oral anticoagulant. Recent data indicate species differences in the expression of CCK receptor subtypes in the pancreas. CCK1 receptors are abundant in rat pancreas but are either absent or present at very low levels in human pancreas. As part of the clinical studies, we examined whether long-term ximelagatran administration causes CCK release and exerts possible trophic effects on the pancreas in humans.
Methods: One hundred thirty patients requiring anticoagulation treatment for atrial fibrillation randomly received, in a double-blind fashion, either 36 mg oral ximelagatran twice daily or warfarin dose adjusted to an international normalized ratio of 2.0 to 3.0. Before enrollment and after 12 months of treatment, computed tomography scans of the pancreas were performed, and pancreas volumes were quantified using the summation-of-areas technique. Three months after the initiation of drug treatment, plasma CCK concentrations were measured by radioimmunoassay 120 minutes after the patients drank 240 mL of a mixed liquid meal (Ensure).
Results: After 3 months of treatment, plasma CCK concentrations did not differ between the ximelagatran and warfarin groups, 15 +/- 18 and 11 +/- 17 pmol/L (X +/- SD; P = 0.22), respectively. The initial average pancreas volumes were 82 +/- 31 and 88 +/- 28 mL in the ximelagatran and warfarin groups, respectively, and decreased to 70 +/- 25 and 75 +/- 28 mL, respectively, after 12 months of treatment. Although the decrease in pancreas volume with time was significant in each group (P = 0.0001), the magnitude of the volume reduction was similar in the 2 groups.
Conclusion: In contrast to rats, in which long-term oral administration of ximelagatran stimulates pancreatic growth and adenoma formation, in humans, ximelagatran does not increase plasma CCK concentrations and has no demonstrable trophic effect on the human pancreas.