High rates of mutation in the TP53 tumor suppressor gene have been found in many human cancers, including breast tumors, making p53 one of the most studied proteins in oncology. However, the prognostic and predictive value of alterations in this gene remains ambiguous. To analyze the clinical value of somatic TP53 mutations, we collected clinical and molecular data on 210 women with primary breast cancer. We found significant associations of p53 mutations with tumor grade, metastasis, molecular subtype, Her2 status and inverse correlations with estrogen and progesterone receptor status. Cox proportional hazard analysis confirmed a strong prognostic value of p53 mutation for overall survival rate and highlighted significant interactions with lymph node involvement and tumor size. In relation to treatment options, TP53 mutations were associated with poor response to anthracyclines and radiotherapy. Categorization of TP53 mutations according to their type and location revealed that patients with nonsense mutation have the poorest prognosis in comparison with wild-type cases and other types of mutations in this gene. Classification of TP53 mutations with respect to the degree of disturbance of protein structure showed association of disruptive mutations with poorer patients' outcome in contrast to wild-type and non-disruptive mutations. In conclusion, the present study confirms p53 as a potential predictive and prognostic factor in oncology practice and highlights the growing evidence that distinct types of mutations have different clinical impacts.