Gene Expression Profiling in Mice can Predict Risk of Skin Lesions Progressing to Cancer
News May 22, 2007
A new study has shown that microarray technology, which allows researchers to simultaneously compare the degree to which hundreds of genes are expressed (converted into proteins), now makes it possible to more definitively identify skin lesions in mice that are thought to be at high risk of progressing to a type of cancer known as squamous cell carcinoma (SCC).
It is hoped that this technique will eventually lead to identification of high-risk SCC tumors in humans. This study appears in the online version of the journal "Oncogene" on May 21, 2007, and was conducted by researchers from the American University of Beirut (Lebanon), Pennsylvania State University, and the National Cancer Institute (NCI), part of the National Institutes of Health.
Researchers compared the gene expression profiles of four different types of mouse skin cells: low-risk and high-risk papillomas (benign tumors) that had been chemically induced, as well as normal skin and SCC.
The investigators demonstrated that precancerous lesions can be separated into subgroups according to distinct patterns of gene activities - namely, which genes were turned on or off. A specific pattern of activity, sometimes called a molecular signature, was present in the precancerous lesions, and correlated with a higher risk for malignant conversion.
"Today's clinicians are in need of a straightforward method of determining which precancerous lesions are at high risk of developing into cancer," said NCI Director John E. Niederhuber, M.D. "The data suggest that if this same correlation holds true for humans, then gene expression analysis of precancerous skin lesions could lead to the development of diagnostic tools for squamous cell carcinoma risk."
Microarrays, also known as gene chips, are glass slides that have been coated with thousands of spots of DNA, each representing a different known gene. Microarray technology is often used in gene expression profiling because it allows researchers to quickly and efficiently identify the multiple genes that are simultaneously expressed in a particular type of cell. Microarray-based comparisons of cancer gene expression patterns in humans have been used to discover subgroups within cancer types as well as to associate specific expression profiles with disease outcomes.
The researchers used microarray technology to test the hypothesis that high-risk papillomas share a similar gene expression pattern with SCC, and therefore represent the likely precursor lesion to SCC. For the analysis, genetic materials which had been isolated from six adult-mouse normal skins, ten low-risk papillomas, ten high-risk papillomas and six SCC were added to the gene chips.
"While we had done previous marker studies to suggest that the high-risk papillomas were the precursors to SCC, the degree of similarity in gene expression patterns between these very early benign lesions and SCC was amazing" said Adam Glick, Ph.D., associate professor, Center for Molecular Toxicology and Carcinogenesis at Pennsylvania State University and formerly of NCI's Laboratory of Cancer Biology and Genetics.
"This suggests that prior to any overt microscopic changes, some lesions have a molecular signature similar, in part, to a malignant tumor."
Several different tests were conducted in this study. One analysis showed that the high-risk papillomas and SCC had very similar molecular signatures or gene patterns, while the low-risk papillomas had very different ones. Another test identified a pattern consisting of 87 genes that distinguished low-risk from high-risk papillomas. This test showed similarities in the expression patterns of the high-risk papillomas and SCC. It also showed that the high-risk papillomas and SCC had reduced expression of genes associated with immune response at an early stage in their development.
A third analysis used the distinguishing set of genes, including the immune function genes, to correctly classify unknown precancerous lesions as low or high risk. Together, these results show that this methodology can predict the risk potential of benign skin lesions with high accuracy in mice. Additional research is needed to see if the genes identified in this study could serve as predictive markers for human cancer.
"In many organs, lesions that turn out to be benign are far more common than those that progress to cancer," said Stuart H. Yuspa, M.D., Chief of NCI's Laboratory of Cancer Biology and Genetics. "Being able to preemptively determine which lesions are likely to progress to cancer provides a strategy both for cancer prevention and for informed follow-up."