Nuclear osteopontin-c is a prognostic breast cancer marker

Background: Although Osteopontin has been known as a marker for cancer progression, the elevated production of this cytokine is not specific for cancer. We have identified the splice variant Osteopontin-c as being absent from healthy tissue but associated with about 75% of breast cancer cases. However, in previous studies of Osteopontin-c, follow-up information was not available. Methods: Here we have analysed 671 patients, comprising a cohort of 291 paraffin blocks plus a population-based case-control study of 380 arrayed breast tumor tissues. Results: We find that high staining intensity of nuclear Osteopontin-c is strongly associated with mortality in patients with early breast cancer. Cytosolic staining for exon 4, reflective of Osteopontin-a and -b also predicts poor outcome. By contrast, total Osteopontin does not correlate with prognosis. These diverse assessments of Osteopontin also do not correlate with each other, suggesting distinct expression patterns for the variant forms. Consistent with its role in tumor progression, not tumor initiation, Osteopontin-c is not correlated with proliferation markers (Ki-67, cyclin A, cyclin B, cyclin E and cyclin D), neither is it correlated with ER, PR or HER2. Conclusions: The addition of Osteopontin-c immunohistochemistry to standard pathology work-ups may have prognostic benefit in early breast cancer diagnosis.