Prognostic significance of the metastasis-associated protein osteopontin in human breast cancer

Abstract

The adhesive glycophosphoprotein (OPN) is capable of inducing metastasis in rodent models ofbreast cancer. We now show that a monoclonal antibody to rat OPN recognizes specifically human OPN using Western blotting techniques andused it to assess the prognostic significance of OPN in primary tumors of a group of 333 patients treated between 1976 and 1982 for operable stage I and stage II breast cancer. The antibody stains immunocytochemically normal breast tissue weakly but pregnant/lactating tissue and 66% of the carcinomas strongly, leaving the remaining 34% as negatively stained. In addition to the carcinoma cells, some host reactive stromal cells, macrophages, lymphocytes, and blood vessels are also stained, but these have been excluded in the following analyses. There is a significant association of staining of carcinomas for OPN with some tumor variables reported previously to be associated with patient outcome: high histological grade (P = 0.024), staining for c-erbB-3 (P < 0.001), p53 (P = 0.014), pS2 (P = 0.025), and borderline significance for progesterone receptor (P = 0.089). The association of staining for OPN with survival times of the patients has been evaluated using life tables over 14-20 years of follow-up (mean 16 years) and analyzed using generalized Wilcoxon statistics. Of the patients who have been classified as OPN-negative, 94% are alive, but only 26% of those classified as OPN-positive are alive after 19 years of follow-up. This association is highly significant (P < 0.0001); the former have a median survival of >228 months and the latter 68 months. When the patients are divided into separate classes based on the percentage of carcinoma cells staining for OPN, the five classes show a progressive decrease in survival with increasing percentage of stained carcinoma cells, and this association is also highly significant (P < 0.0001). Other tumor variables that show a significant association with patient survival times in this group of patients include nodal status, tumor size, histological grade, staining for c-erbB-2, estrogen receptor alpha, or p53. Analysis of the association of patients with carcinomas staining for OPN and their survival in subgroups defined by these tumor variables shows that positive staining for OPN in each subgroup is associated with poorer survival. There is little difference in patient survival times in the OPN-negative group of patients with or without any of the other tumor variables examined. Multivariate regression analysis for 202 patients shows that staining for OPN is most highly correlated with patients' deaths (P < 0.0001), but involved lymph nodes (P = 0.0007), fixed tumors (P = 0.0008), and staining for estrogen receptor alpha (P = 0.008) are also significant independent prognostic variables with that for c-erbB-2 being of borderline significance (P = 0.060). These results suggest that in this group of patients, the presence of the metastasis-associated protein OPN is tightly correlated with patient demise.