Abstract
Immunocytochemical (ICC) detection of disseminated tumor cells (DTC) in bone marrow (BM) in early breast cancer is an independent prognostic factor. The significance of circulating tumor cells (CTC) in peripheral blood (PB) needs further exploration and comparison to DTC detection. PB and BM were prospectively collected from 341 breast cancer patients median 40 months after operation. PB samples were analyzed for tumor cells by a negative immunomagnetic technique (10 × 10^6^ cells/test). BM aspirates were analyzed by standard ICC (2 × 10^6^ cells/test). CTC were present in 10% of the patients and DTC in 14%. Thirty‐seven relapses and 14 breast cancer deaths have occurred at median 66 months after diagnosis. Both CTC‐status and DTC‐status were significantly associated with disease free survival (DFS) (event rate: CTC‐positive 26.5% vs. CTC‐negative 9.1%; DTC‐positive 29.2% vs. DTC‐negative 7.8%) (p < 0.001/p < 0.001, log rank) and breast cancer specific survival (event rate: CTC‐positive 17.6% vs. CTC‐negative 2.6%; DTC‐positive 12.5% vs. DTC‐negative 2.7%) (p < 0.001/p < 0.001). The presence of both CTC and DTC (n = 8) resulted in an especially poor prognosis (p < 0.001). In node negative patients, DTC‐status, but not CTC‐status, predicted differences in DFS (p = 0.006 vs. p = 0.503). Excluding 23 patients with breast cancer‐related events prior to the sample collections, CTC detection was not significantly associated with DFS/distant‐DFS (p = 0.158/0.193), in contrast to DTC detection (p < 0.001/<0.001). Presence of CTC and absence of DTC did not affect DFS (p = 0.516). Applied to early stage disease, CTC analysis of increased volumes of PB appears less sensitive and prognostic than standard DTC analysis. Currently, this does not support an exchange of BM with PB for analysis of occult tumor cells. © 2005 Wiley‐Liss, Inc.