CgA is a tumor marker in NET's (neuroendocrine tumors) but different ranges of sensitivity and specificity according to the commercial assay kits used have been reported. Our aim was to compare three commercial available assay kits that use three different methodologies (IRMA, RIA and ELISA) to determine CgA, in a clinical setting: 52 healthy people, 98 patients with benign diseases, 94 patients with non-NET´s malignancies, 20 SCLC and in 79 patients with NET's.
Results: Using a cut-off with a 100% specificity in healthy people (6 nmol/L, 60 ng/ml, and 90 ng/ml, for RIA, ELISA and IRMA, respectively), abnormal serum concentrations of CgA were found in a high proportion of patients with renal failure (76.7% ,86,7% and 93.3% with ELISA; IRMA and RIA, respectively) other benign diseases (excluding patients with creatinine concentrations > 1.5 mg/dl)(40,3%, 50% and 53,2% with ELISA, IRMA and RIA, respectively) or in patients with non-NET´s malignancies (excluding SCLC and patients with renal failure) (59,8% ELISA, 55,4%% IRMA, 37% RIA). The highest CgA sensitivity in SCLC was obtained with ELISA (100%) and in NET´s with ELISA (83.3%) and IRMA (80.3%) (RIA 65.2%). ROC curves comparing healthy people and NET´s or NET´s- benigns showed a significantly higher area under the curve (AUC) with ELISA (0.964 and 0.774), or IRMA (0.955 and 0.785), and smaller with RIA (0,806 and 0.691).
Conclusions: CgA is not a specific tumormarker and abnormal concentrations may be found in non-NET´s. The higher AUC, sensitivity and specificity obtained with the ELISA and IRMA indicates that these are the best techniques to determine CgA.