Immunomagnetic tumor cell selection - implications for the detection of disseminated cancer cells

BACKGROUND: The optimal method for the detection of disseminated epithelial cancer cells has not yet been found. The standard method, using immunocyto chemistry, offers a sensitivity of up to 10(-6). Molecular methods such as cytokeratin-19 RT-PCR are about 10 times as sensitive, but they are hampere d by interference such as illegitimate gene expression. STUDY DESIGN AND METHODS: Immunomagnetic bead selection of epithelial cance r cells using conjugates directed against the human epithelial antigen (HEA ) followed by immunocytochemistry testing was investigated in this trial. RESULTS: No cytokeratin-positive cells could be enriched from 56 control sa mples. In 104 clinical samples of bone marrow aspirations, PBPC collections , and venous blood obtained from breast cancer patients. the cytokeratin-po sitive rate increased significantly, from 29.9 percent before selection to 54.8 percent after enrichment. Even the yield of detected cancer cells was significantly higher after selection. Up to 2.5 x 10(8) MNCs were easily pr ocessed. However, the mean cancer cell recovery after HEA enrichment was on ly 24.4 percent. Subsequently, selected epithelial cells were successfully immunophenotyped by use of a double-stain technique detecting cytokeratin-p ositive cells and the urokinase-like plasminogen activator receptor. CONCLUSION: HEA bead selection in combination with the standard immunocytoc hemistry method is a powerful and specific tool for the detection of dissem inated cancer cells without false-positive results. Furthermore, it deliver s enough cells for subsequent investigations such as characterization studi es.