Key Gene Found in Chemo-resistant Skin Cancer January 2006

Scientists at the Dana-Farber Cancer Institute and Harvard University have discovered that up to one-fifth of metastasized malignant melanomas harbor a genetic variation, which could provide a potent new diagnostic or drug target for this highly chemotherapy-resistant disease. Malignant melanomas are caused by genetic damage to melanocytes, pigment-producing skin cells responsible for variations in skin color and the color of moles. For patients with malignant melanoma, the prognosis is grim: chemotherapy for late-stage disease invariably fails, and even when treated before metastasis, patients remain at high risk of disease recurrence if the tumor exhibits aggressive histologic features, or if it has spread to the regional lymph nodes. A major hurdle in understanding what makes the cancer so tenacious and chemotherapy-resistant has been discovering the genes that contribute to its development. Dr. William Sellers and colleagues at the Dana-Farber Cancer Institute and Harvard University have done exactly that, publishing their study in the July 2005 issue of Nature. The group used a combination of Affymetrix genotyping and expression microarrays and found that 15 to 20 percent of human metastatic melanomas contained extra copies of a gene called MITF (micropthalmia-associated transcription factor); non-cancerous cells always had the normal number of MITF genes—two, one from the mother and one from the father. The cancer gene turned out to be critical as melanoma patients with extra copies of MITF showed a decrease in five year survival. Sellers' team also discovered that chemotherapy drugs killed melanoma cells more effectively if MITF protein activity was reduced in these cells. Conversely, adding extra copies of the MITF gene to pre-cancerous cells—along with mutations in two other key genes, BRAF and p16—caused those cells to behave like cancer cells in the laboratory. "Reduction of MITF activity may sensitize melanomas to conventional chemotherapeutics," said Sellers. "Targeting MITF may offer a rational therapeutic avenue into this highly chemotherapy-resistant neoplasm." Melanocytes need MITF for normal growth, survival and differentiation. The paradox is that MITF may also help prevent normal melanocytes from developing, causing excessive cell growth properties that might lead to cancer. Scientists know that one of the ways MITF does this is by interacting with the cell cycle inhibitor protein p16; they also know that too much activity from a gene called BRAF may contribute to cancer development, especially if p16 activity is lost. Sellers' team made their discovery using a combination of Affymetrix whole-genome microarrays--genotyping arrays narrowed their search for cancer-causing genes to a small part of the genome, while gene expression array data pinpointed the exact MITF gene. The research group scanned more than 100,000 genetic variations from 58 cancer samples with the GeneChip® Mapping 100K array and found a 3.5 megabase region of the genome—containing approximately 14 genes—that was duplicated (or "amplified") in melanoma samples. They honed in on the MITF gene by analyzing GeneChip® HGU95 expression data for over 12,000 genes, noticing that only one of the 14 duplicated genes was highly expressed in melanoma. The expression microarray data was collected as part of a previous National Cancer Institute experiment that analyzed genome-wide expression for 60 cancer cell lines. Having additional copies of MITF may give malignant melanoma an edge in resisting chemotherapy, but it may also prove to be its Achilles' heel. Developing new drugs to block MITF, or other genes whose function it directs, may allow doctors to effectively treat this disease with conventional or novel therapies. Additionally, now that the mutation is known, physicians could design a genetic screen to identify patients that would benefit most from any type of MITF-focused chemotherapy. Reference: Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma. Garraway LA, Widlund HR, et al. Nature. 436(7047):117-22, July 7, 2005.