Validating gene expression signatures of melanoma

January 1, 2007

The validation of biomarkers is one of the important goals of microarray studies

Noordwijk, Netherlands – Gene expression profiling is considered by most experts in the field to be one of the frontlines in combating melanoma, with recent breakthroughs having the potential to transform current approaches to melanoma diagnosis and prognosis.

University of California, San Francisco (UCSF) researchers were recently able to successfully identify novel biomarkers of melanoma progression, which help us understand better the parameters of radial vs. vertical growth in primary melanoma, as well as the transition from primary to metastatic melanoma, which may ultimately have a crucial influence on melanoma outcome.

Important goals

This can be achieved by conducting such studies at the protein level in independent data sets with the use of immunohistochemistry.

"Recent cDNA microarray analyses of moles, primary and metastatic melanomas performed in our laboratory have identified a large number of genes whose level of expression can be used to distinguish between known stages in the tumor progression cascade of melanoma," says Mohammed Kashani-Sabet, M.D., associate professor of dermatology and co-director of the Melanoma Center at UCSF.

In Dr. Kashani-Sabet's study containing 25 primary cutaneous melanoma specimens, he saw that there was only a loss of gene expression in the transition from radial to vertical growth. He subsequently performed an immunohistochemical analysis of CDH3 and MMP10 in order to confirm that the genes differentially expressed in the radial growth phase by expression profiling were also expressed in the radial growth phase at the protein level.

Results showed that for both markers, the gene expression was significantly higher in the radial growth than in the vertical growth phase. In no case was the expression higher in the vertical than in the radial growth phase, confirming the results obtained by microarray analysis.

"These results suggest the relevance of the differential gene expression signatures found in the various phases of melanoma progression," Dr. Kashani-Sabet tells Dermatology Times.

Using this information, Dr. Kashani-Sabet conducted another gene expression profiling study with 19 metastatic melanomas from 17 patients in order to relate the findings of the vertical growth-specific gene downregulation to the pathogenesis of metastatic melanoma.

"Our results clearly demonstrated that metastatic melanoma is characterized by two different gene expression signatures, with shared features when compared to the signatures of either radial or vertical growth phase primary melanoma. The enormous prognostic potential of gene expression profiles of metastatic melanoma is evident," Dr. Kashani-Sabet says.

Transition

Dr. Kashani-Sabet further examined the transition from primary to metastatic melanoma by using cDNA microarrays.

In his study, he was able to identify 1,024 genes that were differentially expressed in metastatic lesions, when comparing the gene expression signatures of primary vs. unrelated metastatic melanomas.

Interestingly, he noticed that among the genes that were overexpressed in metastatic lesions, he found three clones of nuclear receptor coactivator 3 (NCOA3).

"NCOA3 is a member of the steroid receptor coactivator 1 family, which maps at a region at 20q12 that is frequently amplified in human breast cancers. Intriguingly, increased copy number of 20q is also seen in melanoma samples and cell lines. Our hypothesis was that the primary melanomas expressing high levels of NCOA3 would exhibit an increased risk of metastasis and correspondingly reduced survival, yielding a molecular prognostic marker for melanoma," Dr. Kashani-Sabet says.

Using an immunohistochemical analysis in a melanoma tissue microarray containing primary melanomas from a 343 patient study group, Dr. Kashani-Sabet assessed the prognostic significance of NCOA3 expression.

He found that there was a significant association between NCOA3 overexpression and reduced relapse-free and disease-specific survival (DSS). Results showed that NCOA3 expression was an independent predictor of sentinel lymph node status, and that it was the most powerful factor predicting DSS, outperforming tumor thickness and ulceration.

"It is clear to see from our gene expression profiling analyses, that these novel biomarkers can be used as independent markers of melanoma progression, giving hope for better melanoma outcomes in the future," Dr. Kashani-Sabet explains.

The results of Dr. Kashani-Sabet's studies were also recently published in the Proceedings of the National Academy of Sciences and the Journal of Clinical Oncology.