The utility of circulating levels of human pancreatic polypeptide as a marker for islet cell tumors.

The measurement of plasma levels of human pancreatic polypeptide (hPP) has been reported to be clinically useful in predicting the existence of pancreatic islet cell neoplasms in patients with familial multiple endocrine neoplasia type 1 (FMEN-I) and the possible presence of metastatic disease in patients with islet cell tumors. However, these studies have not been prospective and involve small numbers of patients. In this study, fasting plasma samples from 36 patients with biopsy-proved islet cell tumors were analyzed for hPP by radioimmunoassay and compared with age-matched control subjects. Of 13 patients with FMEN-I who had islet cell tumors, 7 (54%) had elevated plasma hPP levels before surgery. After resection of all islet cell tumors, 4 of 12 patients evaluated after surgery still had elevated levels. Fifteen patients had islet cell tumors that were localized (seven insulinomas and eight gastrinomas), but none of these patients had elevated hPP levels, either before or after surgery. Nine patients, including one with FMEN-I, with metastatic islet cell tumors to the liver were studied; three with more advanced disease had elevated hPP levels before surgery. Each of the nine patients underwent resection of all gross disease and the three patients with elevated preoperative levels had normal postoperative hPP levels. Our results indicate that basal plasma levels of hPP were not clinically useful. The hPP levels did not reliably predict the presence of islet cell tumors in patients with FMEN-I, because 46% of patients with tumors did not have elevated plasma levels, and in those with elevated values hPP levels did not reliably predict the resection of all tumor. Plasma levels of hPP have no utility in patients with localized sporadically occurring islet cell tumors and limited utility (33%) in predicting the presence of metastatic islet cell tumors to the liver.