Malignant melanoma is an extremely aggressive tumor, responsible for 90% of skin cancer mortality. Widespread metastasis is the main cause of death in melanoma patients. Therefore, the identification of molecules related to the degree of melanoma malignancy could be used as prognostic biomarkers. In the present study, using a murine model of melanoma progression, different genes, such as Snai1, Chd1, Gata4, Id4, Cbs, Vav1, and Itga4, were first identified as presenting aberrant DNA methylation in their promoters along melanoma progression by Methylight analysis. The gene expression pattern and epigenetic regulation were validated in these murine cell lines by RT-qPCR and 5AzaCdR treatment, respectively. The genes Angpt2 and Fgf1 were also identified as differentially expressed along tumor progression in this murine model. The expression profile of these genes was also evaluated in primary human melanocytes and RGP (radial growth phase), VGP (vertical growth phase) and metastatic melanoma cell lines. Clinical and gene expression data from 460 patients were obtained from TCGA (The Cancer Genome Atlas). Using an univariate test, patients presenting tumors with low levels of Chd1, Vav1, Id4 and Fgf1, and high levels of Snai1 and Angpt2 were shown to exhibit poor overall survival rate. Indeed, the multivariate analysis using Cox regression model identified high Snai1 (HR=1.217; 95% CI 1.075-1.377; p=0.002) and Angpt2 (HR=1.186; 95% CI 1.059-1.329; p=0.003) expression and low Chd1 (HR=0.757; 95% CI 0.575-0.996; p=0.047), Fgf1 (HR=0.858; 95% CI 0.780-0.944; p=0.002) and Vav1 (HR=0.859; 95% CI 0.764-0.967; p=0.011) expression as independent prognostic factors for the overall survival rate of patients with melanoma. Also using a multivariate analysis, the low levels of Id4 promoter methylation (HR=0.689; 95% CI 0.531-0.891; p=0.045) was shown to be an independent predictive factor for poor survival rate, although no correlation was found between Id4 expression and survival rate. In conclusion, the upregulation of Snai1 and Angpt2, and the downregulation of Chd1, Vav1 and Fgf1 are associated with unfavorable survival outcome, indicating they might be valuable prognostic biomarkers for melanoma. Supported by CNPq and FAPESP.

Citation Format: Alice S. Morais, Ana C. Monteiro, Fernando Andrade, Lucas Goedert, Enilza Espreáfico, Regine Schneider-Stock, André Fujiota, Miriam G. Jasiulionis. Potential biomarkers correlated with poor survival rate of melanoma patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1072.