If this is not your name, click here.
| | If this is not your name, click here. | | |
| | Contact Us | Order Now | Journals | Bookstore | Register a colleague | | |
| |  Gene-Expression Profiling Predicts Tumour Recurrence and Survival for Patients With Adrenocortical Tumours: Presented at ECE By Chris Berrie ISTANBUL, Turkey -- April 29, 2009 -- Routine analysis for the expression of 2 specific 2-gene combinations in patients with adrenocortical tumours can reliably predict the risk of tumour recurrence independent of pathology and can also predict the risk of death independent of tumour stage, according to research presented here at the 11th European Congress of Endocrinology (ECE). Full gene-expression profiling, in addition to discriminating between adenomas and carcinomas, can discriminate 2 types of carcinomas with very different outcomes. At present, a diagnosis of malignant adrenocortical tumour relies on tumour pathology ratings from benign adenoma to malignant carcinoma. Similarly, assessment of prognosis for any given patient relies on tumour staging. The initial question posed by the current study, however, was, "Can gene-expression profiling discriminate benign and malignant tumours?" noted Guillaume Assié, MD, PhD, Cochin Institute, and Endocrinology Service, Assistance Publique Hôpitaux de Paris, Paris, France, speaking here on April 27. Samples from 153 unilateral adrenocortical tumours were used for gene-expression profiling by microarray (n = 92) and/or real-time quantitative polymerase chain reaction (RT-qPCR; n = 148). The initial microarray analysis was carried out by extraction of RNA from the tumour samples and hybridisation to the gene chip for microarray analysis. Unsupervised clustering analyses of the tumour samples provided 2 main groups. When correlated to patient tumour pathology, these provided discrimination of the tumour samples according to benign and malignant pathologies. The addition of patient outcome according to tumour recurrence and patient death demonstrated high significance of gene profiling for prediction of disease-free survival (DFS) (log rank P = 7.1 x 10-10). Within the carcinoma pathology, this analysis also discriminated 2 further groups according to patient survival data alone: bad prognosis and good prognosis (log rank P = 1.2 x 10-4). "Here, in these 2 groups, you can see subgroups of tumours that can only be identified by the gene-expression profile," Dr. Assié noted. A further aim of this research, then, was to build molecular predictors with minimal gene combinations. Expression profiling was used to predict the "top 50 genes" from a training cohort of tumour samples (n = 47) for risk of recurrence (as DFS). These genes were then each analysed by RT-qPCR, with the 2-gene DLG7-PINK1 combination as the best predictor of DFS. The predictive value of this combination was confirmed by an independent validation cohort of the tumour samples (n = 104; log rank P = 1.06 x 10-12). The DLG7-PINK1 combination was compared directly with tumour pathology. Its significance remained when the hazard ratio (HR) of the univariate analysis for the molecular prediction of recurrence was corrected for the Weiss score (P = .01). As Dr Assié added, "We [then] focused on the risk of death using a similar approach." From a training cohort of potentially malignant tumours (Weiss score of 2 or more, n = 23), the BUB1B-PINK1 combination emerged as the best predictor, which was again independently validated (Weiss score of 2 or greater, n = 35; log rank P = 1.68 x 10-6), and which remained an independent predictor after univariate correction for tumour MacFarlane stage (P = 2.0 x 10-5). "Gene-expression profiling provides unique relevant clinical information," Dr. Assié concluded. This information includes not only discrimination of 2 groups of adrenocortical tumours with different outcomes, but also independent prediction of risk of tumour recurrence and risk of patient death using the 2-gene combinations of DLG7-PINK1 and BUB1B-PINK1, respectively. Dr. Assié is one of 6 winners of the European Society of Endocrinology's 2009 Young Investigator Award.
[Presentation title: Gene Expression Profiling Reveals A New Classification of Adrenocortical Tumors and Identifies Molecular Predictors of Malignancy and Survival. Abstract HTC3]
|
