9665 Background: TMAS are a powerful tool for testing a large number of molecular parameters in hundreds of tissue samples. This technique can help us to improve our knowledge of molecular markers used to predict disease evolution.To study clinical factors and the expression pattern of proteins including Rb, p53, ki67, p21, p16, p27, cyclins (A, B1, D1, D3 y E), cyclin dependent kinases (cdks) 1, 2 and 6, Survivin, bcl-2 and activated caspase 3 in 71 patients diagnosed of carcinoma colorectal, of which 31 had metastasic disease at diagnosis 22 relapsed and 18 were of free disease. Patients were included in multicentric clinical trials and had received chemotherapy. Methods: Two TMAS were constructed by acquiring 1 mm cylindrical biopsies from each of formalin-fixed paraffin-embedded tumors and controls and spaced 1.5 mm. As dependent variables we analysed overall survival (OS) and survival after relapsed (SV2). As independent variables we studied age, gender, tumor localization, TNM, lymph node involvement, differentiation grade, liver metastases, symptoms at metastatic disease, and single or multiple metastatic localizations. The univariate analysis was carried out for each one of the variables by means of the Kaplan-Meier method and the multivariate analysis was performed using the Cox regression model. Results: According to the univariate analysis the statistically significant variables were differentiation grade, liver metastasis, symptoms, number localization, cdk1,cdk 2,cdk6, cyclin A, D3, ki-67, p53 and survivin for OS, and differentiation grade, symtoms, number localization, activated caspase 3, cdk2, cdk6, cyclin D3, ki67, p21 and survivin for SV2. In the multivariate analysis those variables that reached a significant value p≤ 0.2 were included. Conclusions: In this study clinical factors such as differentiation grade, symptoms, number localization were independent prognostic factors for OS and symptoms for SV2; none of proteins analysed correlated with OS or SV2 in multivariate analysis. No significant financial relationships to disclose.