A ddress for offprints and correspondence: A. Molino, Divisione Clinicizzata di Oncologia Medica, Piazzale Stefani 1, I-37126 Verona, Italy; Phone 045 8072342; Fax 045 8341277 Breast Cancer Research and Treatment 42: 23–30, 1997.  1997 Kluwer A cademic Publishers. Printed in the Netherlands. Report Bone marrow micrometastases in 109 breast cancer patients: Correlations with clinical and pathological features and prognosis Annamaria Molino 1 , Giuseppe Pelosi 2 , Monica Turazza 1 , Loris Sperotto 1 , Andrea Bonetti 1 , Rolando Nortilli 1 , Giovanni Fattovich 3 , Cristoforo Alaimo 3 , Quirino Piubello 2 , Flavia Pavanel 1 , Rocco Micciolo 4 , and Gian Luigi Cetto 1 Departments of 1 Medical Oncology, University of Verona, 2 Pathology, A zienda Ospedaliera, Verona, 3 Surgery, A zienda Ospedaliera, Verona, 4 Institute of Statistics, University of Trento, Italy Key words: bone marrow, breast cancer, micrometastases, prognosis Summary Background: The presence in bone marrow of cells which react with monoclonal antibodies against tumor- associated antigens has been proposed over the last few years as a new prognostic factor in breast cancer patients. Patients and methods: Bone marrow aspirates were obtained from 109 stage I and II breast cancer patients during or 2–4 weeks after primary surgery. The samples were processed for leukocyte separation on a Ficoll- Hypaque gradient and then used to prepare cytospin slides for immunocytochemical analysis. The slides were stained with a pool of monoclonal antibodies (MoAbs) which recognize tumor associated antigens, using the alkaline phosphatase anti-alkaline phosphatase method. The median follow-up was 36 months (range 15–62); 22 patients relapsed and 7 died. Results: Thirty-four of the 109 patients (31.1%) had MoAb positive bone marrow cells. The bone marrow was positive in 28/74 (37.9%) patients who had the aspirate taken during surgery and in 6/35 (17.1%) who had it taken after surgery (p = 0.055). No association was found between bone marrow positivity and tumour size, nodal status, menopausal status, estrogen receptor positivity or the proliferative index. No association was found between bone marrow and prognosis: the log-rank test was 0.291(p > 0.5) for OS and 0.023 for DFS; the hazard ratio (positive vs negative) was 1.51for OS (95% CI: 0.33–6.86) and 0.93 for DFS (95% CI: 0.35–2.45). Conclusions: In our series, bone marrow positivity did not correlate with prognostic parameters or prognosis. Of interest is the relative excess of positivity when the bone marrow was obtained during surgery. Introduction After primary therapy, 35%–40% of patients with operable breast cancer develop metastases. Adju- vant therapy can improve this prognosis, but it is necessary to identify those patients at risk who are most likely to benefit from this treatment. Stage at diagnosis, as defined by the size of the primary tu- mor and axillary node status, is an important but unsufficient prognostic factor [1] and, in the attempt to identify high risk node-negative patients, many other prognostic factors have been studied (hor- monal receptors, ploidy, proliferative index, etc.). Although bone and bone marrow are frequent sites