Rejection of pancreas allografts is best measured today by co-monitoring the creatinine of a simultaneously transplanted kidney allograft from the same donor. This methodology discourages pancreas transplantation for patients who have previously received a kidney allograft and preuremic patients. Thus, an early, graftspecific marker of rejection is desirable. In this study we compared 2 putative biochemical markers for rejection of pancreas allografts, serum immunoreactive anodal trypsinogen and urinary amylase, with serum creatinine in 15 simultaneously transplanted type I diabetics. Serial values during hospitalizations were determined. Follow-up ranged from 18 to 134 postoperative days. Rejection was diagnosed clinically and considered real if the patient received a course of antirejection medication. Ten of these 15 patients experienced a total of 21 rejection episodes. For all episodes of rejection, serum trypsinogen rose from a baseline of 398.1±25 ng/ml to 1686.2±317.9 ng/ml (PcO.OOl) on the day of rejection. Urinary amylase fell from 88,310± 7877 U/24 hr to 37,508±7142 U/24 hr (PcO.OOl). For 10 patients in whom rejection was diagnosed on the initial hospitalization so that serial prediagnosis sera and urines were available, anodal trypsinogen rose from a baseline of 756±263 ng/ml to 1936±582 ng/ml (-PcO.OOl). Urinary amylase values for these same 10 patients did not change significantly (baseline = 55,788±18,404 U/24 hr, rejection = 47,133+14,737 U/ 24 hr, (P = 0.7). We conclude that serum anodal trypsinogen behaves as a graft-specific biochemical marker for rejection of vascularized pancreas allografts. © 1990 by Williams and Wilkins.