Peripheral Blood Stem Cell Harvests Performed Beyond the Initial Two Days of Apheresis May Not Contribute to Accelerate Granulocyte and Platelet Engraftment After Autologous Stem Cell Transplantation.

Measurement of the number of CD34+ cells in the leukapheresis product is universally used as a surrogate measure of the capacity of peripheral blood stem cells (PBSC) to reconstitute hematopoiesis. Early studies demonstrated that engraftment times are shorter and predictable if at least 5 ×10⁶ CD34+ cells/kg are administered following high-dose chemotherapy. However, satisfactory neutrophil and platelet recovery occurs after infusions of >2.0 ×10⁶ CD34+ cells/kg. Previous analyses demonstrated the presence of different subsets of CD34+ cells in the peripheral blood at different times during mobilization, with pluripotent subsets arising and decreasing earlier after mobilization and more committed subsets peaking later [Stewart et al, Exp Hematol 1995; 23: 1619-27]. We hypothesized that for equivalent CD34+ cell dose, there could be a difference in engraftment kinetics according to the number of days of apheresis required to obtain an adequate PBSC graft.

Data from 270 consecutive autografts performed between July 1999 and April 2006 for non-Hodgkin and Hodgkin lymphoma (58% and 25%, respectively), breast cancer (8%), germ-cell tumors (2%) and acute myeloid leukemia (7%) were analyzed. Mobilization was performed using cyclophosphamide +/- etoposide + G-CSF 10μg/kg; patients received G-CSF after PBSC infusion from day 10 until neutrophil recovery >1500/μL. Patients were stratified in 3 groups according to the SC dose administered as follows: Low-dose (L) <2 ×10⁶ CD34+ cells/kg, Intermediate-dose (I) 2 to 5 ×10⁶ CD34+ cells/kg and High-dose (H) >5 ×10⁶ CD34+ cells/kg and in 2 categories according to the number of days of apheresis performed (≤2 and 3+). The data analysis was divided in 3 steps. Step 1: we made a comparison of engraftment results for group L vs. I and I vs. H. Step 2: to explore potential engraftment differences within each cell dose stratum according to the number of days of apheresis required, that is, I(≤2) vs. I(3+) and H(≤2) vs. H(3+). Step 3: to test the influence of performing additional harvests beyond day 2 on engraftment kinetics we compared group I(≤2) vs. group H(3+), i.e. those who obtained the optimal number of CD34+ cells through additional collections during the same mobilization. Differences between groups were estimated by the Mann-Whitney test and by Cox regression model.

The step 1 analysis confirmed that engraftment was faster for the H group (N=138, median 11 days for ANC>500/μl and N=132, median 11 days for unsupported platelet count >20000/μl) than for the I group (N=118, median 11 days and N=98, 12 days for neutrophils and platelets, respectively) and the L group (N=14, median 12 days and N=12, median 15 days for neutrophils and platelets, respectively). Two-sided p-values favored I in L vs. I (p=0.0185 for neutrophils and p=0.0013 for platelets) and favored H in I vs. H (p=0.0002 for neutrophils and p=0.019 for platelets). The step 2 analysis did not show any statistical difference in engraftment times within groups L, I and H, according to the number of apheresis (≤2 vs. 3+), but there was a trend for faster platelet engraftment for patients in the ≤2 group (p=0.08). There was no difference between the ≤2 and 3+ subgroups in terms of age and number of chemotherapy lines received before transplant; although, in the ≤2 group there were more patients with breast cancer (10.3% vs. 4.8%) and Hodgkin lymphoma (32.2% vs. 16.1%) and fewer pts with non-Hodgkin lymphoma (66.1% vs. 48%) compared to the 3+ group (p<0.001). Step 3 analysis showed no difference for neutrophil recovery between subgroups I≤2 and H3+ (N=28, median 11 days for I≤2 and N=20, median 10 days for H3+, p=0.08) and platelet engraftment (median 11 days for I≤2 and 12 days for H3+, p=0.54). Even dough, patients in the I≤2 cohort received significantly less CD34+ cells/kg (median 4.46, range 2.10-4.99) compared to those in the H3+ group (median 6.33, range 5.10-8.99, p<0.001) and were similar in terms of age and number of lines of prior therapy.

Provided more than 2 ×10⁶ CD34+ cells/kg are obtained in two days of leukapheresis, optimal engraftment times can be achieved even if the threshold of 5 ×10⁶ CD34+ cells/kg is not reached. Our results suggest that the addition of harvest days beyond 2 days to obtain ≥5 ×10⁶ CD34+ cells/kg may not contribute to improved short-term engraftment kinetics.

No relevant conflicts of interest to declare.