OBJECTIVE: To improve our understanding of the molecular mechanisms involved in malignant pheochromocytoma by examining differences in the gene expression profile between benign and malignant tumors. BACKGROUND: The molecular events involved in the malignant transformation of pheochromocytoma are poorly understood. There are also no reliable and uniformly accepted histopathologic criteria to distinguish benign from malignant pheochromocytoma. METHODS: We performed genome-wide expression profiling of 58 pheochromocytomas (29 benign and sporadic, 16 benign and hereditary, 13 malignant) with technical and biologic replication. RESULTS: Unsupervised cluster analysis showed 3 main clusters of tumors that did not have complete concordance with the clinical and pathologic groupings of pheochromocytomas. Supervised cluster analysis showed almost completely separate clustering between benign and malignant tumors. The differentially expressed genes with known function belonged to 8 biologic process categories; signal transduction, transcription, protein transport, protein synthesis, smooth muscle contraction, ion transport, chemotaxis, and electron transport. Gene set enrichment analysis revealed significant correlation between the microarray profiles of malignant pheochromocytomas and several known molecular pathways associated with carcinogenesis and dedifferentiation. Ten differentially expressed genes had high diagnostic accuracy, and 5 of these genes (CFC1, FAM62B, HOMER1, LRRN3, TBX3, ADAMTS) in combination had an area under the receiver operating characteristic (ROC) curve of 0.96 for distinguishing benign versus malignant tumors. CONCLUSIONS: Differentially expressed genes between benign and malignant pheochromocytomas distinguish between these tumors with high diagnostic accuracy. Our findings provide new insight into the genes and molecular pathways that may be involved in malignant pheochromocytomas.
OBJECTIVE: To improve our understanding of the molecular mechanisms involved in malignant pheochromocytoma by examining differences in the gene expression profile between benign and malignant tumors. BACKGROUND: The molecular events involved in the malignant transformation of pheochromocytoma are poorly understood. There are also no reliable and uniformly accepted histopathologic criteria to distinguish benign from malignant pheochromocytoma. METHODS: We performed genome-wide expression profiling of 58 pheochromocytomas (29 benign and sporadic, 16 benign and hereditary, 13 malignant) with technical and biologic replication. RESULTS: Unsupervised cluster analysis showed 3 main clusters of tumors that did not have complete concordance with the clinical and pathologic groupings of pheochromocytomas. Supervised cluster analysis showed almost completely separate clustering between benign and malignant tumors. The differentially expressed genes with known function belonged to 8 biologic process categories; signal transduction, transcription, protein transport, protein synthesis, smooth muscle contraction, ion transport, chemotaxis, and electron transport. Gene set enrichment analysis revealed significant correlation between the microarray profiles of malignant pheochromocytomas and several known molecular pathways associated with carcinogenesis and dedifferentiation. Ten differentially expressed genes had high diagnostic accuracy, and 5 of these genes (CFC1, FAM62B, HOMER1, LRRN3, TBX3, ADAMTS) in combination had an area under the receiver operating characteristic (ROC) curve of 0.96 for distinguishing benign versus malignant tumors. CONCLUSIONS: Differentially expressed genes between benign and malignant pheochromocytomas distinguish between these tumors with high diagnostic accuracy. Our findings provide new insight into the genes and molecular pathways that may be involved in malignant pheochromocytomas.
Authors: Graeme Eisenhofer; Jacques W M Lenders; Gabriele Siegert; Stefan R Bornstein; Peter Friberg; Dragana Milosevic; Massimo Mannelli; W Marston Linehan; Karen Adams; Henri J Timmers; Karel Pacak Journal: Eur J Cancer Date: 2011-10-28 Impact factor: 9.162
Authors: Trudie A Goers; Michael Abdo; Jeffrey F Moley; Brent D Matthews; Mary Quasebarth; L Michael Brunt Journal: Surg Endosc Date: 2012-08-31 Impact factor: 4.584