Does the immunocytochemical detection of epithelial cells in bone marrow (micrometastasis) influence the time to biochemical relapse after radical prostatectomy?

The detection of cytokeratin-positive bone marrow cells has been considered a prognostic factor in numerous malignant tumors. We investigated whether this was also valid for localized prostate cancer. Bone marrow aspirates we re taken prior to radical prostatectomy from 169 consecutive patients with pT1/2 pN0 G1-3 adenocarcinoma of the prostate. The immunocytochemical detec tion of cytokeratin no. 18 (CK 18)positive cells using monoclonal antibody CK 2 was interpreted as micrometastasis. Repeat marrow aspirations were per formed at 6 months postoperatively and once a year thereafter. The patients were re-examined over a period of at least 10 and a maximum of 72 months ( median 32 months). An increase in prostate specific antigen greater than or equal to 0.5 ng/ml was considered a biochemical "relapse". One hundred and fifty-four patients had evaluable bone marrow aspirates, of which 74.7% we re CK 18-negative and 25.3% positive. The latency period for biochemical re lapse was 1481 days (median) in the CK 18-negative group and 1106 days (med ian) in the CK 18-positive group. This difference was not statistically sig nificant. The CK 18-positive aspirates (n = 39) showed one positive cell in 20 cases, two positive cells in 8 and three or more positive cells in 11 c ases. The preoperative number of cells had no statistically significant eff ect upon the onset of biochemical relapse. Only patients with three or more CK 18-positive cells tended to have a poorer prognosis. One hundred and th irteen patients had evaluable bone marrow aspirates pre- and postoperativel y. Postoperative persistence or occurrence of CK 18-positive cells did not affect the outcome of the disease. The detection of CK 18-positive cells in bone marrow does not influence the prognosis of patients with localized pr ostate cancer within a period of 32 months (median). Solely a subgroup of p atients showing a large preoperative number of CK 18-positive cells seems t o tend to an unfavorable course of the disease. Thus, further studies are n ecessary aiming at a more detailed characterization of these cells.