Non-specifically labelled cells that simulate bone marrow metastases in patients with non-metastatic breast cancer.

Aims: To determine whether the presence of disseminated bone marrow tumour cells at diagnosis is a prognostic factor for breast cancer patients at high risk of recurrence or bone metastasis, and to assess their presence as a criterion for evaluation of the potential benefits of adjuvant chemotherapy.

Methods: Multiple bone marrow aspirates from 72 breast cancer patients free from metastasis were obtained during surgery at the time of diagnosis and were tested immunologically by alkaline phosphatase antialkaline phosphatase technique with a panel of three antiepithelial monoclonal antibodies (MoAb) KL1, EMA, and HMFG2.

Results: In nine of 72 patients, with each MoAb tested, numerous strongly positive cells always isolated were observed. However, it was demonstrated that these cells were non-specifically labelled and could be found in normal controls.

Conclusions: There was no evidence of marrow tumour cells in 72 operable breast cancer patients. It is suggested that published results may be greatly overestimated and that non-specific labelling may be undetected. More specific MoAb should be found and a correlation with molecular biology should be performed if this criterion is to be considered as a prognostic factor.