PURPOSE: We evaluated the pattern of chromogranin A (CgA) plasma levels in a large number of patients with neuroendocrine tumors (NETs), in a series of patients with chronic atrophic gastritis (CAG) with and without enterochromaffin-like (ECL) cell hyperplasia, and in healthy participants (HPs). PATIENTS AND METHODS: Two hundred thirty-eight patients with NETs, 42 patients with CAG with or without ECL cell hyperplasia, and 48 HPs were studied. All patients underwent a baseline visit, biochemical routine check-up, imaging techniques, endoscopy, and histologic determination. RESULTS: CgA plasma levels were higher in patients with NETs compared with CAG patients or HPs (P < .001). In the NET group, we observed higher CgA levels in patients with diffuse disease compared with patients with local or hepatic disease (P < .001). CgA plasma levels were significantly higher in patients with Zollinger-Ellison syndrome compared with other types of endocrine tumors (P < .001). We found the best cutoff range between HPs and NET patients to be 18 to 19 U/L (sensitivity, 85.3%; specificity, 95.8%). Comparing all participants without neoplasia (HPs, CAG patients, and disease-free patients) and patients with endocrine tumors, the best cutoff range was 31 to 32 U/L (sensitivity, 75.3%; specificity, 84.2%). Setting the specificity at 95%, the cutoff range was 84 to 87 U/L (sensitivity, 55%). CONCLUSION: Our study confirms the high specificity and sensitivity of CgA in diagnosing an endocrine tumor. It is necessary to use a cutoff range of 84 to 87 U/L to obtain a high specificity in diagnosing NETs, with the aim of excluding patients in whom the CgA was elevated as a result of other non-neoplastic diseases.
PURPOSE: We evaluated the pattern of chromogranin A (CgA) plasma levels in a large number of patients with neuroendocrine tumors (NETs), in a series of patients with chronic atrophic gastritis (CAG) with and without enterochromaffin-like (ECL) cell hyperplasia, and in healthy participants (HPs). PATIENTS AND METHODS: Two hundred thirty-eight patients with NETs, 42 patients with CAG with or without ECL cell hyperplasia, and 48 HPs were studied. All patients underwent a baseline visit, biochemical routine check-up, imaging techniques, endoscopy, and histologic determination. RESULTS:CgA plasma levels were higher in patients with NETs compared with CAG patients or HPs (P < .001). In the NET group, we observed higher CgA levels in patients with diffuse disease compared with patients with local or hepatic disease (P < .001). CgA plasma levels were significantly higher in patients with Zollinger-Ellison syndrome compared with other types of endocrine tumors (P < .001). We found the best cutoff range between HPs and NET patients to be 18 to 19 U/L (sensitivity, 85.3%; specificity, 95.8%). Comparing all participants without neoplasia (HPs, CAG patients, and disease-free patients) and patients with endocrine tumors, the best cutoff range was 31 to 32 U/L (sensitivity, 75.3%; specificity, 84.2%). Setting the specificity at 95%, the cutoff range was 84 to 87 U/L (sensitivity, 55%). CONCLUSION: Our study confirms the high specificity and sensitivity of CgA in diagnosing an endocrine tumor. It is necessary to use a cutoff range of 84 to 87 U/L to obtain a high specificity in diagnosing NETs, with the aim of excluding patients in whom the CgA was elevated as a result of other non-neoplastic diseases.