Molecular characterization of melanoma cancer stem cells

Melanoma that has spread to distant sites is rarely curable with standard therapy, and most patients with distant metastasis are appropriately considered candidates for clinical trials exploring new forms of treatment (Stewart, World Cancer Report, IARC Press, 2003). The objective response rate to dacarbazine and the nitrosoureas, carmustine and lomustine is approximately 10%. A recent phase III trial comparing fotemustine and dacarbazine showed higher overall response rate (15% vs 8%) in the fotemustine arm compared to the dacarbazine arm in first-line treatment of disseminated melanoma (Avril et al, JCO 2004). Responses are usually short-lived, ranging from 3 to 6 months, although long-term remissions can occur in a limited number of patients who attain a complete response. The molecular mechanisms underpinning the differences between responding and non-responding patients are not known to date. Recent evidence has demonstrated that cancers can be viewed as an abnormal organ in which tumor growth is driven by a population of cancer stem cells (CSCs), which can give rise to both more CSCs as well as non-tumorigenic cancer cells (Jordan et al., NEJM 2006). In marked contrast to the CSCs, these latter cells have either no or a markedly diminished capacity to form new tumors. This observation has implications for the biology of tumor formation as well as the diagnosis and treatment of cancer. It can be hypothesized that the high percentage of failure observed in melanoma treatments might be due to differences in the chemosensitivity of CSCs present in malignant melanoma. To treat cancer effectively, the CSCs must be eliminated. Otherwise, the tumor will rapidly reform if the therapy eliminates nontumorigenic cancer cells but spares a significant population of the CSCs. Indeed, CSC can not only be the source of all the malignant cells in the primary tumor but also compose a small reservoir of drug resistant cells that are responsible for relapse after remission induced by chemotherapy or generate distant metastases. The existence of CSC has been shown in various types of leukemia, in breast cancer and in nervous system tumors (gliomas, Jordan et al., NEJM 2006). Interestingly, CSCs have been identified in melanoma too: Fang and collaborators identified a stem cell-like population in human metastatic melanomas propagating as nonadherent spheres when cultured in growth medium suitable for human embryonic stem cells (Fang et al. Cancer Res. 2005). Individual cells from melanoma spheres could differentiate under appropriate conditions into multiple cell lineages, recapitulating the plasticity of neural crest stem cells. Multipotent melanoma spheroid cells persisted after serial cloning in vitro and transplantation in vivo, indicating their ability to self-renew. Most significantly, melanoma CSC (mCSC) were more tumorigenic than adherent cells when grafted to mice. Moreover, it has been shown that mCSC could also be isolated from metastatic melanoma cell lines (Grichnik et al JID 2006). Preliminary data from Unit3 confirmed these data and showed that mCSC could be isolated from melanoma lung metastases as well cell cultures of both metastatic and primary melanoma. The aim of this project is characterizing the genomic, gene expression, proteomic and miRNA profiles of mCSCs. We will attempt to causally correlate the molecular characteristics of mCSC with their biological properties, tumorigenicity, metastatic potential and resistance to chemotherapic drugs.