137 Hematopoietic Cell Yield Declines Predictably Over Time During Apheresis

Track: Contributed Abstracts
Wednesday, February 13, 2013, 6:45 PM-7:45 PM
Hall 1 (Salt Palace Convention Center)
Allison F Hazlett, MS , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA
Elaine Wang , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA
Jennifer Anderson , Pathology, VCU Massey Cancer Center, Richmond, VA
Harold Chung, MD , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA
John Michael McCarty, MD , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA
William Clark, MD, MS , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA
Catherine H. Roberts, PhD , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA
Amir Ahmed Toor, MD , Bone Marrow Transplant, VCU Massey Cancer Center, Richmond, VA

Hematopoietic cell (HC) mobilization to support high dose therapy is generally carried out using cytokines with or without chemotherapy. The resulting HC yield and the duration of cell collection have been well studied and tend to be variable. No consistent unifying relationship has been described which would reliably represent the kinetics of this process between different patients. To accomplish this, the HC yield from the mobilization procedures for 431 patients was examined.  The diagnoses were multiple myeloma (N=220), non-Hodgkin's lymphoma (155) and Hodgkin's lymphoma (56). Mobilizing regimens included chemotherapy + GCSF (97), GCSF (232), GCSF + plerixafor (84). To normalize the HC yield between patients, the total number of CD34+ cells collected on a given day was divided by the volume of blood (L) processed and was termed the HC mobilization index (HCMI). For the combined cohort the mean HCMI value on day 1 of apheresis (HCMI1) was 19.7 (± 38.9) × 106 CD34+ cells/L/day. A significant correlation was found between HCMI1 and the circulating CD34+ cell count on day 1 (R2 0.69, p<0.01), and the total HC yield in each patient (0.97, p<0.01). These observations were consistent in patients with various diagnoses and receiving different mobilizing regimens. Daily HCMI values were plotted over days of apheresis to determine the rate of change. A general trend of declining HCMI values over days of apheresis was observed, with some patients showing an initial increase. To offset the effect of the large range of HCMI values observed, the logarithm (log) of HCMI for each day was plotted against day of apheresis for each patient with >2 days of collection (n=279) to give individual log-HCMI decay curves. A quadratic equation (y=ax2+bx+c) provided the best fit for these curves (mean R2 0.88), demonstrating a parabolic relationship such that log HCMI increased and declined in proportion to square of time (in days) following the start of apheresis. It was noted that the values of coefficients were normally distributed in the study population; coefficient a (mean, -0.04 ± 0.2), b (0.04 ± 1.1), and c (1.7 ± 1.2). A Two-step cluster analysis of these coefficients distinguished three main groups (clusters) of patients.  Patients in cluster 1 (38%) tended to have an initial increase in HCMI followed by an asymptotic decline; those in clusters 2 (50.2%) and 3 (8.6%) displayed declines of varying magnitude. Cluster 1 showed a higher proportion of NHL patients (p=0.01), whereas MM was over-represented in cluster 2 (p=0.008), consistent with the notion that prior chemotherapy received by NHL patients impacted their mobilization kinetics.  In conclusion, we demonstrate that HC yield during various mobilization procedures follows mathematically predictable kinetics which may reflect underlying hematopoietic reserve.