Limited Diagnostic Utility of Chromogranin A Measurements in Workup of Neuroendocrine Tumors
Jonas Baekdal, Jesper Krogh, Marianne Klose, Pernille Holmager, Seppo W. Langer, Peter Oturai, Andreas Kjær, Birgitte Federspiel, Linda Hilsted, Jens F. Rehfeld, Ulrich Knigge, Mikkel Andreassen
Abstract
BACKGROUND: Plasma chromogranin A (CgA) is related to tumor burden and recommended in the follow-up of patients diagnosed with neuroendocrine tumors (NETs). The use of CgA in the workup of a suspected NET is more questionable. OBJECTIVE: To assess the positive predictive value (PPV) of CgA plasma concentrations above the upper reference limit (URL) in patients with suspected NET. METHOD: Patients referred to the NET Centre, Rigshospitalet, Copenhagen from 2015 to 2019 with clinically suspected NET were included if a CgA measurement was performed prior to referral. The utility of CgA was assessed by comparing pre-referral CgA concentrations to the outcome of a thorough workup. In 47 selected cases with continuously unexplained elevated CgA concentrations, a processing-independent analysis (PIA) for CgA was performed. RESULTS: A total of 197 patients were included. NET was ultimately diagnosed in 25 patients. CgA plasma concentrations were above the URL (elevated) in 19/25 patients diagnosed with NET. In total, 167/197 had elevated CgA concentrations at referral. The positive predictive value (PPV) of elevated CgA concentration was 11% (19/167). Proton pump inhibitor (PPI) treatment was identified as the possible cause of CgA elevation in 55/148 patients with falsely elevated CgA. CgA concentration was normal in 28/47 patients when using PIA. CONCLUSION: Our data do not support using measurement of CgA for screening when NET is suspected since the PPV was rather low. PPI treatment is a common cause of increased CgA concentrations and should always be discontinued before CgA measurement. PIA of CgA could be a way of excluding NET when suspicion is based primarily on elevated CgA.