CliniMACS CD34-selected cells to support multiple cycles of high-dose therapy.

Abstract

Background: Traditionally, following high-dose therapy (HDT), unmanipulated autologous PBPC are infused. Alternatively, purified CD34+ cells can now be obtained by immunomagnetic separation using the CliniMACS device. Limited data currently exist examining hemopoietic recovery with such cells. Method(s): Ten patients with advanced breast cancer had PBPC mobilized with docetaxel (100 mg/m2) and G-CSF (10 mug/kg per day), harvested and processed using the CliniMACS CD34-selection device and equally divided into three aliquots for cryopreservation. Unmanipulated 'back-up' cells were also collected on a separate day of the same mobilization, divided into three and cryopreserved. Patients subsequently received three cycles of HDT with cyclophosphamide (4 g/m2), thiotepa (300 mg/m2) and paclitaxel (175 mg/m2). The intent was for patients to receive CD34-selected cells to support each of the three cycles of HDT (i.e., 1/3 for each cycle). If, however, hemopoietic recovery was delayed after Cycle 1, 1/3 of the unmanipulated cells were infused following Cycle 2 and the remaining CD34-selected cells (2/3) were used to support Cycle 3. Result(s): PBPC from 10 patients underwent CD34-selection with a resulting median purity of 93% (range: 76-98%) and yield of 62% (range: 16-93%). Of the 10 patients, only two were able to be supported with CD34-selected cells for all three cycles of HDT. The remaining eight patients required unmanipulated 'back-up' cells to support Cycle 2. Three patients also required infusion of 'back-up' unmanipulated cells because of persistent neutropenia (n = 1) or thrombocytopenia (n = 2) following cycles initially supported by CD34-selected cells. The median number of CD34-selected cells (x 106/kg) infused per cycle was 1.5 (0.7-2.6) (n = 20) and unselected cells was 1.7 (1.4-2.8) (n = 10). Comparing hemopoietic recovery between cycles of HDT supported by CD34-selected (n = 20) and unmanipulated cells (n = 10) there was a significant slowing with the CD34-selected cells; time to ANC > 1.0 = 13 days versus 10 days, platelets > 20 = 17 days versus 13 days, > 50 = 25 versus 17 days (all P values < 0.001). There was no correlation between the dose of CD34-selected cells infused and neutrophil/platelet recovery. Discussion(s): We have demonstrated that, although unmanipulated PBPC achieve rapid hemopoietic recovery (at modest CD34 doses of <= 2.8 x 106/kg), CliniMACS-selected CD34+ cells (in the doses utilized in this study of <= 2.6 x 106/kg) result in significantly prolonged recovery.