Utility of osteopontin as a biomarker in recurrent epithelial ovarian cancer

Objective. Osteopontin (OPN) is overexpressed in tumors and serum of ovarian cancer patients and may serve as a biomarker. To evaluate the utility of serum osteopontin in monitoring disease status, we evaluated 234 serum samples from post-oophorectomy patients with ovarian cancer and 38 samples from healthy controls. Methods. Serum samples were collected from 203 women with recurrent ovarian cancer and 31 newly diagnosed women participating in an experimental chemotherapeutic clinical trial. Controls included I I young healthy women and 27 peri- or postmenopausal women without ovarian cancer. Samples were assayed for osteopontin using an enzyme-linked immunosorbent assay (ELISA) kit. Statistical analyses for group comparisons of biomarker distribution used the nonparametric Wilcoxon's rank sum test for two-group comparisons and the Kruskal-Wallis test for three-group comparisons. Results. Osteopontin values ranged from 25 to 1463 ng/ml for patients and 25 to 617 ng/ml for controls. Mean patient levels were lower than mean control levels (74 ng/ml vs. 147 ng/ml, respectively, P = 0.0006). Serum osteopontin levels correlated with recurrent disease versus remission (68 ng/ml vs. 34 ng/ml, P = 0.0034), presence of ascites versus absence (71 ng/ml vs. 53 ng/ml, P = 0.0002), and bulky disease vs. nonbulky disease (75 ng/ml vs. 38 ng/ml, P = 0.0005). CA-125 values yielded the same trends with greater statistical difference. Conclusions. These results demonstrate that serum osteopontin concentrations in post-oophorectomy patients with recurrent ovarian cancer are not greater than in healthy controls. Nevertheless, within this heterogeneous patient population, the values do correlate with bulk of disease. The potential utility of this assay in monitoring women with CA-125 negative disease is worthy of exploration. (C) 2004 Elsevier Inc. All rights reserved.