Chromogranin A as serum marker of pituitary adenomas

OBJECTIVE The diagnostic impact of chromogranin A (CgA) measurement has been studied in various neuroendocrine tumours (NET) such as pheochromocytomas, gastrinomas and neuroblastomas. Clinically nonfunctioning pituitary adenomas (NFPA) are generally diagnosed on tumoural symptoms or hypopituitarism and, except for gonadotrophins and their free subunits which may be increased in the case of gonadotrophinomas, markers of endocrine secretory activity are lacking not only for diagnostic purpose but also in the postoperative follow-up of these patients. As the presence of CgA has been demonstrated by immunohistochemistry in pituitary adenomas, we performed this study to further assess the sensitivity of CgA measurement in sporadic pituitary adenomas using a new, specific, sandwich immunoassay. SUBJECTS We first completed a basal normative data set obtained using this assay by studying four healthy men (49 +/- 13 years old), five healthy premenopausal women (35.8 +/- 7.5 years old) and five healthy postmenopausal women (49.1 +/- 4.6 years old) basally and after TRH administration. Twenty-seven patients [12 men (64.2 +/- 11.8 years), even premenopausal women (38.4 +/- 5.7 years) and eight postmenopausal women (67.7 +/- 10.3 years)] with NFPA, 15 acromegalic patients [nine men (45 +/- 13.3 years), six women (52 +/- 14.9 years)] and 19 patients with a prolactin-secreting adenoma [four men (41.2 +/- 18 years) and 15 women (31.2 +/- 7.5 years), with a macroadenoma (n = 11) or a microadenoma (n = 8)] had basal and TRH-stimulated measurement of CgA. A gonadotrophin-releasing hormone (GnRH)-stimulation test was also performed in two, four and four patients, respectively. All patients had sporadic pituitary adenomas. MEASAUREMENTS Serum CgA was measured using a solid-phase two-site immunoradiometric assay based on monoclonal antibodies that bind to two distinct contiguous epitopes within the 145-245 region of CgA. RESULTS Mean basal CgA concentration in 14 normal subjects was 80.2 ng/ml (SD: 31.7; range 19-124). A cut-off value for normal range was thus set at 125 ng/ml. TRH injection did not change significantly the CgA levels, peak values remaining less than 124 ng/ml. Three out of 27 subjects with NFPA (11%) had elevated basal CgA levels (576, 143, 241 ng/ml, respectively). Serum levels of CgA were not influenced by TRH in any of the NFPA subjects (including those three with increased basal levels). One out of 15 acromegalic patients (6.6%) and one out of 19 hyperprolactinemic patients (5.2%) had elevated serum basal CgA which did not significantly increase after TRH administration. In the remaining patients TRH-tests did not modify CgA levels. GnRH administration did not modify CgA levels. CONCLUSIONS CgA serum levels measurement, assessed with a novel assay, does not provide a helpful marker for the clinical management of functioning and NFPA.