Identification of candidate genes associated with salivary adenoid cystic carcinomas using combined comparative genomic hybridization and oligonucleotide microarray analyses.

Adenoid cystic carcinoma (ACC) of the salivary gland often has a variable clinical course with a poor prognosis. To investigate DNA copy number aberrations associated with ACCs, we compared comparative genome hybridization data from ACCs (n = 6) with other types of salivary gland tumors such as adenocarcinomas (n = 3) and pleomorphic adenomas (n = 6). While 15 gain loci (1q32, 6p25, 6q21-q24, 7q11.2, 7q31, 10q11.2, 11p12-q12, 12q13, 12q14, 13q24, 16p13.3-13.2, 18p11.3, 18q23, 19q13.4, and Xq28) were detected, no DNA loss locus was evident. To examine the expression status of genes on the ACC-associated loci, transcriptional measurements of approximately 38000 human genes then were monitored using Affymetrix U133 Plus 2.0 GeneChips. A total of 4431 genes were found differentially expressed by at least two-fold between ACCs and normal salivary glands. Of them, 3162 genes were up-regulated and 1269 genes were down-regulated in ACCs. After obtaining locus information about the RNA transcripts from the Affymetrix database, we found 262 ACC-associated genes with increased expression on ACC-associated loci. To investigate functional network and gene ontology, the 262 genes were analyzed using Ingenuity Pathway Analysis Tool. The function with the highest P value was a cancer-related function (P = 2.52e-4 to 4.71e-2). In addition, we identified pituitary tumor-transforming gene 1 and transformation related protein 63 genes that were up-regulated by increasing DNA copy number and modulated expression of oncogenes. These results suggested that the combination of copy number and gene expression profiling provides an improved strategy for gene identification in salivary gland ACCs.