Abstract 4230


The optimal injection site for cytokine administration when used to mobilize peripheral blood stem cells for collection is unclear. There are known differences in the pharmacokinetics of subcutaneously injected drugs based upon site adiposity. We hypothesized that injection in lower adipose-tissue-containing sites in the extremities would result in a reduced reservoir effect leading to lower exposures of granulocyte colony stimulating factor (G-CSF) and therefore reduced stem cell collection following cytokine mobilization.


We completed a prospective single institution IRB-approved randomized study to determine the efficiency and tolerability of different injection sites among patients with multiple myeloma or lymphoma undergoing stem cell mobilization and apheresis. The primary end-points were the total number of CD34+ cells collected and the number of days of apheresis required to collect target numbers (5 × 10E6 CD34+ cells/kg for patients with lymphoma; 10 × 10E6 CD34+ cells/kg for patients with myeloma). Forty patients were randomized to receive cytokine injections in their abdomen (group A) or extremities (group B). Randomization was stratified based upon diagnosis (myeloma; N=29 vs. lymphoma; N=11), age (≤50; N=13 vs. >50; N=27), and mobilization strategy (cytokines alone; N=27 vs. chemomobilization; N=13). Both group A and B were balanced with respect to the stratification criteria. Filgrastim was planned at a dose of 10 ug/kg/day for patients undergoing chemomobilization or 15 ug/kg/day for patients undergoing cytokine-only mobilization. Actual mean cytokine doses were 11.78 ug/kg/day using chemomobilization and 12.96 ug/kg/day using cytokines alone due to rounding to nearest vial size. Patients recorded the injection site for G-CSF and symptoms daily.


Of those enrolled, 90% were evaluable with 18 patients in each group. Four were deemed non-evaluable due to failure to proceed to the planned mobilization procedure (1 in group A and 2 in group B) or lack of consistent injection site (1 patient). In addition, one patient in group B received a non-protocol specified injection of plerixafor due to poor mobilization and collected a total of 13.62 × 10E6 CD34+ cells/kg in 2 days of apheresis. Among the 36 evaluable subjects, 1 subject in each group failed collection with a total of < 2.0 × 10E6 CD34+ cells/kg collected. Mean BMI at the time of mobilization was not different between groups A and B (27.25 ± 4.7 versus 29.39 ± 5.7, respectively; p=NS). Mean numbers of CD34+ cells (±SD) collected were not different between groups A and B (9.15 ± 4.7 versus 9.85 ± 5 × 106/kg, respectively; p=NS). The mode and median duration of apheresis was 2 days for both groups. Subjects from both groups reported similar toxicities of pain and discomfort at the injection site.


Based upon the analysis, G-CSF administration site (extremities versus abdomen), does not affect the number of CD34+ cells collected by apheresis or the duration of apheresis needed to reach the target cell dose.


Lonial:Celgene: Consultancy; Millennium: Consultancy, Research Funding; BMS: Consultancy; Novartis: Consultancy; Gloucester: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.