Classification of Follicular Thyroid Tumors by Molecular Signature

• Published in: Clinical cancer research Vol. 9; no. 5; p. 1792
• Main Authors: Catherine B. Barden, Katherine W. Shister, Baixin Zhu, Gerardo Guiter, David Y. Greenblatt, Martha A. Zeiger, Thomas J. Fahey III
• Format: Journal Article
• Language: English
• Published: American Association for Cancer Research 01-05-2003
• ISSN: 1078-0432; 1557-3265

Purpose: Thyroid nodules are common, with a lifetime risk of developing a clinically significant thyroid nodule of 10% or higher.
Preoperative diagnosis was greatly enhanced by the introduction of fine needle aspiration in the 1970s, but there has been
little advancement since that time. Discrimination between benign and malignant follicular neoplasms is currently not possible
by fine needle aspiration and can even be difficult after full pathologic review. The purpose of these studies is to identify
genes expressed in follicular adenomas and carcinomas of the thyroid that will permit molecular differentiation of these neoplasms.
Experimental Design: Gene expression patterns of 17 thyroid follicular tumors were analyzed by oligonucleotide array analysis. Gene profiles for
follicular adenomas and carcinomas were identified, and the two groups were compared for differences in expression levels.
The differentially expressed genes were used to perform a hierarchical clustering analysis training set. Five follicular tumors
with diagnosis undisclosed to the investigators and 2 minimally invasive carcinomas were entered into the cluster analysis
as a test set to determine whether diagnosis by gene profile correlated with that obtained by pathologic evaluation.
Results: Thyroid follicular adenomas and carcinomas showed strikingly distinct gene expression patterns. The expression patterns of
105 genes were found to be significantly different between follicular adenoma and carcinoma. Many uncharacterized genes contributed
to the distinction between tumor types. For five follicular tumors for which the final diagnosis was undisclosed, the clustering
algorithm gave the correct diagnosis in all 5 cases.
Conclusions: Gene profiling is a useful tool to predict the molecular diagnosis of follicular thyroid tumors. Genes were identified that
reliably differentiate follicular thyroid carcinoma from adenoma. This study provides insight into genes that may be important
in the molecular pathogenesis of follicular thyroid tumors, as well candidates for preoperative diagnosis of follicular thyroid
carcinoma.