A Higher Number Of CD34+ Cells Collected During Mobilization Is Independently Associated With Successful Engraftment In Autologous Stem Cell Transplant Patients

Interactions between bone marrow stromal cells and hematopoietic stem cells (HSC) that allow mobilization and subsequent peripheral blood stem cell (PBSC) collection for autologous bone marrow transplantation (ASCT) are similar to interactions that allow successful engraftment after reinfusion. While it has been demonstrated that the cell dose of reinfused CD34+ cells, a surrogate marker for the HSC dose, is an important predictor for engraftment in ASCT, there is little information about the influence of the number of mobilized CD34+ cells on engraftment. We hypothesized that the number of mobilized CD34+ cells can be an indicator for the quality of the surface interactions between HSC and stromal cells and therefore allow prediction of homing potential. We performed a retrospective analysis on patients who underwent PBSC collection and subsequent ASCT at our institution to assess whether efficiency of PBSC mobilization is predictive of engraftment failure.

We identified 369 patients who underwent PBSC collection between 01/01/2006-8/31/2012 for a first ASCT at Montefiore Medical Center. We collected data on patient age, sex, use of lenalidomide or thalidomide (“Imid”) prior to mobilization, mobilization regimen (GCSF, GCSF +mozobil, chemo+GCSF, chemo+GCSF+mozobil), number of collections for final cell dose, number of CD34+ cells infused, and the presence of a positive blood culture within 30 days of ASCT. Quintiles were created for the number of CD34+ cells collected. The primary outcome was engraftment failure defined as not achieving an absolute neutrophil count (ANC) >1000/mL or a platelet count >50,000/mL (no platelet transfusion in </= 7 days) by day 30 post-ASCT. Secondary outcomes were time to ANC and platelet engraftment (number of days to achieve an ANC>500 and a platelet count >20,000/mL without transfusion within </=7 days). We performed a multivariate logistic regression analysis to assess the association of collected CD34+ cells and engraftment failure while adjusting for the other variables. For time to event analyses we used Cox proportional hazard models.

Median age at ASCT was 58 (range 19-82). 56% of patients were male. Patient-reported race was as follows: Black (38%), White (17%), and “Other” (45%). Indications for ASCT were Multiple Myeloma (45%), Non-Hodgkin Lymphoma (41%), Acute Leukemia (9%), Hodgkin Lymphoma (3%), Amyloidosis (1%), and Germ Cell Tumors (1%). The median number of CD34+ cells collected was 7.7x10 ^6/kg (range 2.26-120 x10 ^6/kg) and median number of CD34+ cells infused was 5.3x10 ^6/kg (2.3-45x10 ^6/kg). Median number of collections for transplant dose was 2 (range 1-8). CD34 cells collected were divided into quintiles (cut points: 6.04, 7.57, 9.86 and 17.7x10 ^6/kg)

We found that a higher number of collected CD34+ cells during mobilization was associated with less engraftment failure (p=0.0067): every increase in quintile was associated with a 40% decrease in the risk of engraftment failure (OR 0.60, 95%CI 0.41-0.87). A higher number of collected CD34+ cells was associated with decreased time to platelet engraftment (HR1.15, CI 1.00-1.32, p=0.052), but not with time to ANC engraftment (HR 1.07, p=0.35). Additionally, positive blood cultures within 30 days of ASCT were associated with engraftment failure (p=0.0035), while race, sex, number of collections for the transplanted dose and mobilization regimen did not appear to affect engraftment. Interestingly, we observed that prior Imid use demonstrated a trend toward less engraftment failure (OR 0.41, 95%CI 0.17-1.01; p=0.052).

Although a moderate correlation was observed between the variables CD34 cells collected and CD34 cells infused, a sensitivity analysis did not identify a significant difference in model estimations when either variable was omitted. Therefore, it is unlikely that coefficients and variance estimates were affected significantly by the observed collinearity between CD34 cells collected and infused. A larger study is needed to verify these findings, as the impact of collinearity often diminishes with sample size.

The number of CD34+ cells collected independently predicts successful engraftment and Imid use prior to mobilization may protect against engraftment failure. A possible explanation for these findings is that alterations in the bone marrow environment influence mobilization and later engraftment of HSC.

Barta:Onyx: Membership on an entity’s Board of Directors or advisory committees, Speakers Bureau; Janssen Pharmaceuticals: Speakers Bureau; Seattle Genetics: Membership on an entity’s Board of Directors or advisory committees; Otsuka: Research Funding.