www.fondation-medic.ch

Our group studies prognostic and predictive factors in melanoma based upon present understanding of melanoma biology. While prognostic parameters (Breslow thickness, Ulceration, Number of Mitosis, ...) are well defined on the melanoma primary lesion (Disease Stages I or II), only very few are used in clinical assessment at later stages.
We studied a melanosomal protein, Tyrp1, in metastatic tissue and found that high expression of the marker strongly correlated with a short survival. Interestingly, expression of the marker also correlated with Breslow thickness, the most important prognostic parameter in the primary lesion. This finding suggests a possibly important role of this protein in melanoma progression and aggressiveness and that it might be used as a target for therapy due to its contant expression regardless of the stage of disease. Studies of Tryp1 are further pursued.

About 50% of melanomas bear a mutated BRAF gene. BRAF mutated melanoma cells paradoxically exihibit a senescence-like phenotype and a low proliferation index compared to wild type. The consequence of such mutation is a constant activation of one of the major pathways in the melanocyte: the MAPK/MEK/ERK pathway. Specific inhibition of this pathway in these patients by newly designed drugs (e.g. Vemurafinib) resulted in a spectacular regression of the metastatic lesions in about half of the treated patients. However, the duration of the response is limited and resistance to the treatment appears in all cases. Contrasting with these important observations, we found in a panel of melanoma cell lines that BRAF mutations clearly drive a cell senescence-like phenotype (Fig.1) associated with low proliferation. Most interestingly, these characteristics can be reversed in some melanoma cell lines by inhibitors of mutated BRAF or MEK. These data suggest that these cells are less aggressive than the wild type ones and contrast with the status of “driver mutation” known in other tumors.

In situ detection of β-galactosidase activity in wild type and V600EBRAF melanoma cells.

Mutated cKit has been an indication for the use of specific inhibitors in mucosal melanoma where this mutation is rather frequent. The frequent activation of SRC in melanoma has also been a target for therapy. Both strategies led to limited success.

We conducted a screening in a panel of cell lines with three important mutations known in melanoma (BRAF, NRAS and cKit) as to their sensitivity to cKit and SRC inhibitors. We found a group of lines extremely sensitive to the drug dasatinib with inhibitory concentrations (IC50) as low as 10-10M while other lines ranged from moderately sensitive to resistant (IC50=10-8-10-4M) (Table 1). We could identify common features explaining this high sensitivity: All highly sensitive cells are BRAF and NRAS wild type, they express high levels of cKit but no correlation with SRC expression. Dasatinib treatment resulted in a dramatically inhibition of cKit, SRC, ERK and AKT phosphorylation in sensitive cells, while it had no effect on the phosphorylation of both ERK and AKT in the mutated ones, suggesting a selective effect on proliferation/survival of cells with cKIT expression not harbouring NRAS/BRAF mutation.

This has an interesting clinical relevance since about 25% of metastatic melanoma patients (wild type for BRAF and NRAS) would potentially benefit from dasatinib treatment considering the wide therapeutic window of the drug in this particular group.