The tyrosine kinase membrane receptor HER-2 has usually been associated to a poor outcome and with an increased rate of relapse in breast cancer patients. However, the different studies about this receptor have been controversial and at the present moment its expression is considered a predictive factor related with medical treatment response. In our study we have characterized the expression of the complete HER-2 receptor, p185HER-2, and a truncated one, with a molecular weight of 95kDa, generated by a proteolytic shedding of the complete form, called p95HER-2. Our results suggested that the expression of this truncated forma, p95HER-2, was significantly more frequent in breast cancer samples of patients with an higher number of nodal metastasis, being its expression more frequent in this tissues that in primary breast tumors. Therefore, p95HER-2 could be a marker or cause of local metastasis, the most important prognostic factor of poor outcome in breast cancer. Our results suggested that an higher p95HER-2 expression level was directly related with a poor outcome and an higher risk of relapse. In concrete, the median disease free survival (DFS) of the patients which samples showed a low p95HER-2 expression was of 139 months whereas DFS of those with breast cancer samples overexpressing p95HER-2 was of 39 months. In fact, 63 % of patients with high p95HER-2 expression tumors suffered a relapse when we considered 5 years DFS. The statistical analysis showed that p95HER-2 was an independent prognostic factor then, its expression could be useful to detect those patients with an higher risk of relapse that could be consequently treated with the optimal preventive treatment. The p95HER-2 activation was also significantly more frequent in breast cancer samples of patients with an higher number of metastatic nodes and more advanced stage. Its phosphorylation was also significantly related with MAPK and Akt activation. In fact, those patients with a simultaneous p95HER-2 and MAPK or p95HER-2 and Akt activation showed a significantly shorter DFS, identifying a patient group with more
