Malignant pheochromocytoma - Chromaffin granule transmitters and response to treatment

Chromaffin granule transmitters such as chromogranin A and catecholamines have been used in the diagnosis of pheochromocytoma, but the diagnostic and prognostic value of chromogranin A have not been explored in malignant pheochromocytoma. We evaluated these transmitters in patients with pheochromocytoma (n = 27), both benign (n = 13) and malignant (n = 14). Patients with benign pheochromocytoma were studied before and after surgical excision (n = 6), whereas patients with malignant pheochromocytoma were evaluated before and after combination chemotherapy with regular cycles of cyclophosphamide/dacarbazine/vincristine (nonrandomized trial in n = 9). During treatment, patient responses to chemotherapy were divided according to anatomic and clinical criteria: responders (n = 5) versus nonresponders (n = 4). Plasma chromogranin A rose progressively (P < 0.0001) from control subjects (48.0 +/- 3.0 ng/mL) to benign pheochromocytoma (188 +/- 40.5 ng/mL) to malignant pheochromocytoma (2932 +/- 960 ng/mL). Parallel changes were seen for plasma norepinephrine (P < 0.0001), though plasma epinephrine was actually lower in malignant than benign pheochromocytoma (P = 0.0182). In bivariate analyses, chromogranin A, norepinephrine, and epinephrine discriminated between pheochromocytoma and control subjects tall P < 0.0001), whereas in a multivariate analyses, norepinephrine was the best discriminator (P = 0.011). Chromogranin A was significantly different in benign versus malignant pheochromocytoma on both bivariate (P = 0.0003) and multivariate (P = 0.011) analyses. After excision of benign pheochromocytoma, chromogranin A (P = 0.028), norepinephrine (P = 0.047), and epinephrine (P = 0.037) all fell to values near normal. During chemotherapy of malignant :pheochromocytoma (n = 9), plasma chromogranin A (P = 0.047) and norepinephrine (P = 0.02) fell but not epinephrine. In 5 responders to chemotherapy, there were significant declines in chromogranin A (P = 0.03) and norepinephrine (P = 0.03) but not epinephrine; in 4 nonresponders, none of the transmitters changed. Plasma chromogranin A varied longitudinally with tumor response and relapse. We conclude that plasma chromogranin A is an effective tool in the diagnosis of pheochromocytoma, and markedly elevated chromogranin A may point to malignant pheochromocytoma. During chemotherapy of malignant pheochromocytoma, chromogranin A can be used to gauge tumor response and relapse.