MUNICHThe presence of bone marrow micrometastases in breast
cancer patients, as determined by detection of cytokeratin-positive
cells, predicts a poor prognosis regardless of lymph node status,
according to a German study.
Bone marrow micrometastases correlated significantly with an
increased risk of distant relapse after treatment but not with
locoregional relapse. The correlation was evident in patients with
and without lymph node involvement, although patients who had both
nodal involvement and micrometastases in the bone marrow had the
worst prognosis of all.
It was interesting to find that survival was similar between
patients who had lymph node metastases but no bone marrow metastases
and those who had no lymph node metastases but did have bone marrow
involvement, Stephan Braun, MD, of Ludwig-Maximilians
University, Munich, said at the San Antonio Breast Cancer Symposium.
The results were recently published in the New England Journal of Medicine
The findings came from an analysis of bone marrow aspirates obtained
from 552 patients with stage I-III breast cancer and 191 control
patients with nonmalignant conditions. The aspirates were evaluated
by means of an assay using the monoclonal antibody A45-B/B3, which is
directed against an epitope common to several cytokeratin polypeptides.
The analysis revealed tumor cells in bone marrow samples from 199
(36%) of the 552 patients with breast cancer vs 2 (1%) of the 191
controls. The presence of tumor cells correlated with inflammatory
breast cancer, histologic involvement of 10 or more lymph nodes, and
higher grade of the primary tumor. Bone marrow micrometastases did
not correlate with lymph node metastases per se, Dr. Braun said.
At a median follow-up of 38 months, the presence of bone marrow
micrometastases correlated significantly with distant metastases: 79
events in patients with bone marrow involvement vs 28 in patients
without micrometastases in the marrow (P < .001). No
association was seen between bone marrow micrometastases and
The presence of bone marrow micrometastases also adversely affected
survival, Dr. Braun said. Of the patients with bone marrow
micrometastases, 25% died of cancer-related causes during follow-up
vs 6% of those without micrometastases. Overall survival was 68% in
patients who had micrometastases in the marrow, compared with 93% in
patients without bone marrow micrometastases (P < .0001).
A multivariate analysis that included lymph node status, hormone
receptor status, and bone marrow status showed that the presence of
bone marrow micrometastases had the highest relative risk for
mortality (2.51-fold increased relative risk of cancer-related death).
In a separate analysis, Dr. Braun and his colleagues evaluated 245
patients who had small primary node-negative tumors (less than 2 cm).
The analysis showed that among patients who did not receive systemic
adjuvant therapy, those who had bone marrow micrometastases had a
lower survival than patients who did not have bone marrow micrometastases.
These results show that bone marrow micrometastases have an
independent prognostic effect on the early onset of distant
metastases and on decreased survival in patients who have stage I-III
breast cancer, Dr. Braun said. Perhaps more importantly,
this immunoassay might be useful to monitor complementary treatment
strategies aimed at preventing metastatic disease.