The aim of this study was to identify genes that are predictive of carcinogenic change in patients of oral leukoplakia (OLK) using a cancer-related microarray. Candidate biomarkers were discovered using the Oligo GEArray OHS-802 and validated on independent samples by semi-quantitative reverse transcription-PCR (RT-PCR) and real-time RT-PCR. Both the discovery and the validation cohorts of samples included normal oral tissues, OLK tissues and oral squamous cell carcinoma (OSCC) tissues. Based on the microarray results, we found that there were nine genes successively up-regulated or down-regulated more than 2-fold between the normal group and OLK group and then again between the OLK group and OSCC group. The expression levels of the nine signature genes had statistically significant differences (p<0.05) between the OLK and normal groups and between the OSCC and OLK groups. In summary, the expression of the 9 signature genes might be representative of OLK carcinogenesis. A cancer-related microarray was used to identify a panel of candidate biomarkers for determining carcinogenesis of OLK lesions, in combination with semi-quantitative and real-time RT-PCR to validate the results. Our data indicated that alterations in gene expression that result in carcinogenesis can be identified in precancerous oral tissues.