Peripheral blood progenitor cells (PBPCs) are increasingly used for autografting after high-dose chemotherapy. One advantage of PBPCs over the use of autologous bone marrow would be a reduced risk of tumor-cell contamination. However, the actual level of tumor cells contaminating PBPC harvests is poorly investigated. It is currently not known whether mobilization of PBPCs might also result in mobilization of tumor cells. We evaluated 358 peripheral blood samples from 46 patients with stage IV or high-risk stage II/III breast cancer, small cell (SCLC) or non- small cell (NSCLC) lung cancer, as well as other advanced malignancies for the detection of epithelial tumor cells. Monoclonal antibodies against acidic and basic cytokeratin components and epithelial antigens (HEA) were used in an alkaline phosphatase-anti-alkaline phosphatase assay with a sensitivity of 1 tumor cell within 4 x 10(5) total cells. Before initiation of PBPC mobilization, circulating tumor cells were detected in 2/7 (29%) patients with stage IV breast cancer and in 2/10 (20%) patients with extensive-disease SCLC, respectively. In these patients, an even higher number of circulating tumor cells was detected after chemotherapy with VP16, ifosfamide, and cisplatin (VIP) followed by granulocyte colony-stimulating factor (G-CSF). This approach has previously been shown to be highly effective in mobilizing PBPCs. In the 42 patients without circulating tumor cells during steady state, tumor cells were mobilized in 9/42 (21%) patients after VIP+G-CSF induced recruitment of PBPCs. The overall incidence of tumor cells varied between 4 and 5,600 per 1.6 x 10(6) mononuclear cells analyzed. All stage IV breast cancer patients and 50% of SCLC patients were found to concomitantly mobilize tumor cells and PBPCs. Kinetic analyses showed two patterns of tumor cell recruitment depending on the presence or absence of bone marrow disease: (1) early after chemotherapy (between days 1 and 7) in patients without marrow infiltration, and (2) between days 9 and 16 in patients with marrow infiltration, ie, within the optimal time period for the collection of PBPCs. We show that there is a high proportion of patients with circulating tumor cells under steady-state conditions, and in addition a substantial risk of concomitant tumor cell recruitment upon mobilization of PBPCs, particularly in stage IV breast cancer patients with bone marrow infiltration. The biologic and clinical significance of this finding is unknown at present.

Peripheral blood progenitor cells (PBPCs) are increasingly used for autografting after high-dose chemotherapy. One advantage of PBPCs over the use of autologous bone marrow would be a reduced risk of tumor-cell contamination. However, the actual level of tumor cells contaminating PBPC harvests is poorly investigated. It is currently not known whether mobilization of PBPCs might also result in mobilization of tumor cells. We evaluated 358 peripheral blood samples from 46 patients with stage IV or high-risk stage II/III breast cancer, small cell (SCLC) or non- small cell (NSCLC) lung cancer, as well as other advanced malignancies for the detection of epithelial tumor cells. Monoclonal antibodies against acidic and basic cytokeratin components and epithelial antigens (HEA) were used in an alkaline phosphatase-anti-alkaline phosphatase assay with a sensitivity of 1 tumor cell within 4 x 10(5) total cells. Before initiation of PBPC mobilization, circulating tumor cells were detected in 2/7 (29%) patients with stage IV breast cancer and in 2/10 (20%) patients with extensive-disease SCLC, respectively. In these patients, an even higher number of circulating tumor cells was detected after chemotherapy with VP16, ifosfamide, and cisplatin (VIP) followed by granulocyte colony-stimulating factor (G-CSF). This approach has previously been shown to be highly effective in mobilizing PBPCs. In the 42 patients without circulating tumor cells during steady state, tumor cells were mobilized in 9/42 (21%) patients after VIP+G-CSF induced recruitment of PBPCs. The overall incidence of tumor cells varied between 4 and 5,600 per 1.6 x 10(6) mononuclear cells analyzed. All stage IV breast cancer patients and 50% of SCLC patients were found to concomitantly mobilize tumor cells and PBPCs. Kinetic analyses showed two patterns of tumor cell recruitment depending on the presence or absence of bone marrow disease: (1) early after chemotherapy (between days 1 and 7) in patients without marrow infiltration, and (2) between days 9 and 16 in patients with marrow infiltration, ie, within the optimal time period for the collection of PBPCs. We show that there is a high proportion of patients with circulating tumor cells under steady-state conditions, and in addition a substantial risk of concomitant tumor cell recruitment upon mobilization of PBPCs, particularly in stage IV breast cancer patients with bone marrow infiltration. The biologic and clinical significance of this finding is unknown at present.