The presence of immunoreactive prostate-specific antigen (IR-PSA) has been reported in breast cancers and has been suggested to confer a positive prognosis. However, recent large, well-controlled studies have found no significant prognostic value when IR-PSA positivity is examined as an independent variable, even when ultrasensitive immunofluorometric techniques are utilized. The present study, using indirect immunohistochemistry on 75 formalin-fixed, paraffin-embedded breast cancers shows PSA immunoreactivity in only seven of 75 cases (9%), suggesting that PSA positivity in breast carcinoma is not useful as a prognostic or tumor marker with hospital-based methods.