Astrocytomas are the most common primary tumors of the central nervous system. According to the WHO classification, there are four grades of astrocytoma, considering their histology, molecular characteristics and prognosis, where the grade IV (glioblastoma) is the deadliest form and still without effective therapy. The aim of this dissertation was to investigate changes in gene structure and protein expression of Dishevelled family and transcription factors TCF1 and LEF1, the molecular components of the Wnt signaling pathway, in astrocytic brain tumors and compare results with clinical parameters. Four selected microsatellite markers for DVL1 (D1S468 and D1S243), DVL2 (D17S960) and DVL3 (D3S1262) genes have shown a significant number of samples demonstrating MSI and LOH, genetic changes that contribute to the genome instability in cancer. Diffuse astrocytomas (WHO grade II) analyzed with D1S468 (p=0.008) have shown the highest percentage of MSI (53.8%), while LOH was the most common in glioblastoma (31.2%) analysed with D3S1262 (p=0.007). The results obtained on the total sample indicate that the microsatellite instability is constantly present, with a more frequent appearance in lower grades and may be the cause of astrocytoma formation, whereas large deletions are significantly associated with the highest grade and have a role in progression. The levels of DVL1, DVL2 and DVL3 protein expression did not show statistically significant correlation with genetic changes. Nevertheless, statistically significant differences in the number of cells with certain level of expression of a single protein between the grades were observed. Bivariate correlation of all analyzed proteins showed a statistically significant positive correlation between DVL3 and TCF1 (p=0.020), DVL3 and LEF1 (p=0.006), TCF1 and LEF1 (p=0.021), while DVL1 and DVL3 were negatively correlated (p=0.002). DVL1 was also negatively correlated (p<0.001) with astrocytoma grade, while DVL3 (p<0.001), TCF1 (p=0.008) and LEF1 (p<0.001) showed positive correlation with malignancy grade suggesting involvement of these proteins in malignant progression. These results contribute to a better understanding of the molecular profile of astrocytic brain tumors and could serve as molecular markers of progression.