IMMUNOASSAYS FOR MEASUREMENT OF CHROMOGRANIN-A AND PANCREASTATIN-LIKEIMMUNOREACTIVITY IN HUMANS - CORRESPONDENCE IN PATIENTS WITH NEUROENDOCRINE NEOPLASIA

Chromogranin A (CgA) is a useful marker of neuroendocrine tumors in hu mans. Here we describe and compare two immunoassay methods for determi nation of CgA, a radioimmunoassay (RIA) and an enzyme linked immunoass ay (ELISA). The detection limit of the ELISA was lower than that of th e RIA method (2 ng/ml versus 10 ng/ml, respectively), though the CgA R IA method covered a wider range than the CgA ELISA (10-920 ng/ml versu s 2-500 ng/ml, respectively). There was no cross-reactivity with synth etic human and porcine pancreastatin (PST) in the two assays. There wa s a significant positive correlation between levels of CgA in sera fro m patients with carcinoid disease, measured by the two methods (r = 0. 9, p < 0.0001), and the values were in the same range. Similarly, seru m CgA levels in normal controls were also in the same range when assay ed by the two methods. A commercially available porcine PST RIA method was evaluated, especially with respect to the influence of Sep-Pak ex traction of serum on the levels of pancreastatin-like immunoreactivity (PST-LI). Ten sera from carcinoid patients were treated with Sep-Pak extraction, and levels of PST-LI were determined in non-extracted and extracted sera. There was a significant positive correlation between t he concentrations of PST-LI measured in extracted and non-extracted ca rcinoid sera (r = 0.9, p < 0.002), and the levels were in the same ran ge. There was also a significant positive correlation between levels o f CgA and PST-LI in 49 carcinoid sera (r = 0.8, p< 0.0001). In the PST RIA there was a minimal cross-reactivity with human CgA (0.08%). In c onclusion, we have found that the two immunoassays for measurement of CgA are sensitive and reproducible, and give similar results. Extracti on of serum previous to assaying PST-LI seems not to be necessary.