Chromogranin A storage and secretion: sensitivity and specificity for the diagnosis of pheochromocytoma.

Chromogranin A, co-stored and co-released with catecholamines from adrenal medullary and sympathetic neuronal vesicles, is elevated in the plasma of patients with pheochromocytoma. The usefulness of the hormone in the differential diagnosis of hypertension is examined. An elevated level of chromogranin A had comparable diagnostic sensitivity (83%, 24/29) to, but greater diagnostic specificity (96%, 86/90) than the level of plasma catecholamines when subjects with pheochromocytoma (n = 29) were evaluated in comparison to several reference groups, including normotensive controls (n = 49), subjects with essential hypertension (n = 28), subjects with renovascular hypertension (n = 5), and subjects with primary aldosteronism (n = 3). Subjects with signs or symptoms suggesting pheochromocytoma, but in whom the diagnosis was ultimately ruled out (n = 5) had normal plasma levels of chromogranin A. A modest rise in chromogranin A in those with essential hypertension, and correlation of chromogranin A with diastolic blood pressure in normotensive patients and patients with essential hypertension did not impair the diagnostic usefulness of chromogranin A for pheochromocytoma. Renal failure was associated with an elevated plasma chromogranin A independently of blood pressure. Plasma chromogranin A correlated with tumor mass, tumor chromogranin A content, tumor norepinephrine content, and urinary vanillylmandelic acid excretion; it did not correlate with plasma or urinary catecholamines, nor with blood pressure in patients with pheochromocytoma. Plasma chromogranin A levels did not differ in subjects with pheochromocytoma when stratified by age, sex, tumor location, or tumor pathology. Several drugs used in the diagnosis or treatment of pheochromocytoma (clonidine, metoprolol, phentolamine, and tyramine) had little effect on plasma chromogranin A concentration. Within the pheochromocytoma, chromogranin A was localized along with catecholamines to the soluble core of chromaffin granules, where it accounted for 18 +/- 5% of vesicle soluble protein. We conclude that 1) chromogranin A emerges along with catecholamines from pheochromocytoma chromaffin granules; 2) plasma chromogranin A is a sensitive and specific diagnostic tool in evaluation of actual or suspected pheochromocytoma; 3) plasma chromogranin A predicts pheochromocytoma tumor size and overall catecholamine production; and 4) drugs commonly employed in the diagnosis or treatment of pheochromocytoma have little effect on plasma chromogranin A level, preserving the usefulness of chromogranin A in evaluating pheochromocytoma. Thus, measurement of chromogranin A provides a useful adjunct to the diagnosis of pheochromocytoma.