The rise and fall of chromogranin A as an indicator for NET

#53

Introduction: There is now general awareness that treatment with proton pump inhibitors (ppi) may cause moderate rises in circulating concentrations of gastrin and/or chromogranin A (CgA), thus decreasing the specificity of these markers in the diagnosis of neuroendocrine tumors (NETs). Due to their high efficacy, proton pump inhibitors (ppi) are one of the most frequently prescribed classes of drugs worldwide. Several studies in various countries report that 30% or more of in-patients have been prescribed ppi. Patients who are suspected of having a neuroendocrine tumor (NET) often have gastrointestinal (GI) symptoms for which ppi are prescribed and therefore they may be already taking ppi when their first blood sample is assayed for NET markers. We previously encountered a case of extreme rise in CgA which flagged the possibility of a NET, but was later shown to be due to ppi therapy alone. We present CgA and gastrin data in relation to ppi therapy and withdrawal for this patient. In order to see if this was an isolated incident, we audited all laboratory requests for CgA/gastrin in a two-year period and monitored those that provided details of ppi treatment/withdrawal.

Aim(s): The audit was undertaken to evaluate the number of request forms that provided information on ppi therapy. Additionally all CgA and gastrin results on samples received during the audit period were analyzed to assess the specificity of these tests for NET. The specific case follow-up allowed us to relate CgA and gastrin levels to repeated initiation and withdrawal of ppi therapy.

Materials and methods: Samples were measured for gastrin by our own 'in-house' radioimmunoassay and chromogranin A was measured by ELISA (Dako Ltd.)

Conference: 7th Annual ENETSConcerence (2010)

Presenting Author:

Authors: Armstrong L, Ryan K, Ardill J,

Keywords: Chromogranin A, proton pump inhibitors, gastrin, audit, case report,

To read the full abstract, please log into your ENETS Member account.