Collection of Autologous Peripheral Blood Stem Cells Using the COBE Spectra Is Associated with Higher CD34+ Yields and Requires Less Collection Time Than the Fresenius AS104 Cell Separator.

We retrospectively analyzed yields and characteristics of autologous peripheral blood stem cell (PBSC) products collected using either a COBE Spectra (CS) or Fresenius AS104 (F) cell separator. CS and F collections were performed according to manufacturer's protocol.

One hundred and fifty nine adult patients (pts) collected between January 2006 and July 2009 at our center were randomly selected for analysis. Eighty nine pts had multiple myeloma (MM) and 72 pts other cancers, including non-Hodgkin's lymphoma (n=42), Hodgkin's disease (n=20), testicular cancer (n=5), neuroblastoma (n=1), acute lymphocytic leukemia (n=1), or acute promyelocytic leukemia (n=1). Pts were collected on either CS (n=85) or F (n=74), and mobilized with G-CSF alone (n=47), G-CSF and Plerixafor (P) (n=26), or G-CSF and Chemotherapy (Ch) (n=66). Data from pts who failed to mobilize were not included in the analysis. Results are summarized in table. All p-values were calculated using two-tailed, heteroscedastic student T-test; statistical significance was defined as p < 0.05.

CD34 yields are calculated as the ratio of the absolute CD34 count/μl of the PBSC product to the pt's peripheral blood CD34 / μl count taken on the day of collection. Pt age, weight on day of collection, number of collections per pt, volume processed per collection, disease type (MM versus other) and mobilization strategy did not differ between patients collected with CS or F. Products collected on the CS had significantly higher total WBC when compared to the F (163.3 ± 136 K/μl vs. 55.7 ± 29.2 K/μl, p < 0.001), but had lower percentages of mononuclear cells (75% ± 23% vs. 85% ± 11%, p < 0.001); this relationship persisted when patients were divided by disease type or mobilization strategy. CD34 / kg counts of PBSC products were not significantly different in any cohort, but absolute CD34 count of the PBSC product was significantly higher in every disease type and mobilization strategy on the CS (all pts, 727 ± 1326 cells/μl vs. 265 ± 789 cells/μl, p < 0.001). CD34 yields were also significantly higher on the CS than the F across all disease types and mobilization strategies (all pts, 24.9 ± 10.9 vs. 11.2 ± 7.4, p < 0.001). Analysis of peripheral blood CD34 counts/ μl and product CD34 count/ μl showed a stronger correlation in pts mobilized with G-CSF + P (R² = 0.8735) versus G-CSF + Ch (R² = 0.5492) or G-CSF alone (R² = 0.5). Hours for collection were lower on the CS than the F, regardless of disease type (4.25 ± 0.65 hrs vs. 5.2 ± 0.98 hrs, p < 0.001). No difference in WBC or platelet engraftment after transplantation was observed between the cell separators. Adverse effects on infusion were similar except, 3 of 169 pts experienced seizures upon infusion of PBSC product; all three were collected on the CS and all three had product WBC counts of 590 K/μl or above (compared to a median of 163.3 K/μl for the total MM/CS population).

In conclusion, collection of PBSC using the CS is associated with higher CD34 yields and requires less collection time than collections with the F cell separator.

No relevant conflicts of interest to declare.