The aim of the present study was to evaluate the clinical usefulness of the cytokeratin tumor marker M3/M21 as a screening, prognostic, and monitoring marker for ovarian cancer and as a predictive marker in patients with adnexal masses. In order to determine the specificity of the M3/M21 test we investigated M3/M21 serum levels in several benign conditions. The cytokeratin tumor markers M3/M21 and Tissue Polypeptide Specific Antigen (TPS) were also investigated in the follow-up of ovarian cancer patients. We evaluated M3/M21 serum levels in 75 patients suffering from ovarian cancer FIGO stages Ia to III, using a prototype immunoradiometric assay (IRMA). Sera of patients with benign cysts, endometriosis, pelvic inflammatory disease, inflammatory bowel disease and liver cirrhosis were evaluated in 90, 10, 38, 10, and 20 cases, respectively. Furthermore, we analyzed TPS serum levels by means of IRMA during the follow-up of 40 patients suffering from ovarian cancer. With a sensitivity of 57% and a specificity of 95% M3/M21 was not suitable as a screening marker for ovarian cancer. Although M3/M21 was able to discriminate between ovarian cancer and benign adnexal tumors (univariate logistic regression, p = 0.0003), M3/M21 did not provide additional information (in addition to CA 125) (multivariate logistic regression, p = 0.2). M3/M21 serum levels were elevated in several benign conditions such as liver cirrhosis and inflammatory bowel disease. In ovarian cancer patients elevated M3/M21 serum levels prior to therapy were associated with a poor overall and disease-free survival (log-rank test, p = 0.03, and log-rank test, p = 0.01, respectively). In patients with recurrent ovarian cancer M3/M21 and TPS showed median lead-time effects of 3.2 and 3.9 months, respectively. M3/M21, while obviously not suitable for screening or differential diagnosis of adnexal masses, could be useful as an additional prognostic factor. M3/M21 and TPS are valuable tumor markers in the follow-up of ovarian cancer patients.