Expression data from head and neck squamous cell carcinoma

Forty-four paired (from the same patient) samples of head and neck squamous cell carcinoma (HNSCC) and normal tissue were studied with Affymetrix U95A chips. A stringent multi-test approach, combining 7 traditional and microarray-specific statistical tests, was used to analyze the resultant data. Candidate genes were assigned to tiers of significance based on the number of statistical tests that each gene satisfied. Representative genes (both up-regulated and down-regulated) from each of the 3 tiers would be quantified with RT-PCR on both microarray-tested and new samples of HNSCC. The goal of this study is to identify reliable differentially-expressed genes on HNSCC and to testify our hypothesis whether or not a combinatorial approach (multi-tests) to analyzing microarray data can really identify differentially-expressed genes with fewer false-positives. Keywords: disease state; tumor vs. normal Paired HNSCC tumor and normal samples from 22 patients were evaluated for differential gene expression on Affymetrix U95A chips.