gms | German Medical Science

Background: Although only less than 10% of women with primary breast cancer have clinicopathologic signs of overt metastases, metastatic relapse occurs in about half of the cases with apparently localized tumors within five years after surgery due to hematogeneous and/or lymphatic spread of occult tumor cells. During the last 7 years, we studied 398 patients with non-metastatic breast cancer for the detection of cytokeratin (CK)-positive cells in the bone marrow (BM). We now evaluated the prognostic value of these results and compared these data with tumor cell spread to the lymph nodes with regard to clinical outcome.

Methods: From each patient, bilateral BM (10 ml) aspirations were evaluated seperately, analyzing 6x106 BM cells by immunocytochemistry (IC) using the pan anti-CK antibody A45-B/B3.

Results: The overall CK-positive rate was 41% (164/398 patients), 44% (81/184) for patients with small tumor size (< 2 cm), 41% (103/250) for node-negative patients and 46% (19/41) for patients with G1 tumors, respectively. Analyzing both BM aspirates separately, we found that 67% of the CK+ patients were only positive in one of the two aspirates studied. After a 7 year follow-up time, 46/398 (12%) patients died, 25/398 (6%) showed a relapse and 327/398 (82%) are still in complete remission (CR). In those patients who relapsed or died (n=71), no tumor cell spread was found in 21/71 (21%) patients. In 35/71 (49%) patients, tumor cells mostly were found either in the lymph nodes or in the BM but rarely in both compartments (15/71; 21%). In the latter group, hematogeneous and lymphatic spread correlated with T2/T3 tumors and a higher number of involved lymph nodes. In CR patients, no tumor cell spread was found in 126/327 (39%) patients and in 46/327 (14%) patients tumor cells were found in the lymph nodes and the BM. In 155/327 (47%) tumor cells were detected either in the lymph nodes or in the BM.

Conclusions: Our data show that there is a frequent hematogeneous or lymphatic spread of tumor cells from very small breast tumors with comparable prognostic value. To identify BM-positive patients, we recommend the analysis of two BM aspirations to avoid false-negative results.