Abstract
Abstract 2336
Poster Board II-313
The principal aim of our study was to determine the level of CXCR4 expression in lymphocyte cells and CD34+ mobilized cells, SDF1 polymorphism and SDF-1 plasma levels in CLL. The secondary objective was to check if there is a correlation of these variables with PBSC mobilisation, disease characteristics and outcome. We studied 73 CLL patients enrolled in SFGM-TC & FCLLG phase III randomized trial. Patients received 3 mini-CHOP followed by PBSC mobilisation and then 3 courses of Fludarabine. Patients were randomised between autologous HSCT versus “watch and wait” for responders and between autologous HSCT versus Fluda+cyclophosphamide in non-responders after 2 DHAP. After induction, there were 20 CR, 28 PR, 10 SD and 5 PD. Thirty-two patients were mobilized, the median number of collected CD34+ cells was 3.5×106/kg [0-53.6] with a median of 15% of lymphocyte infiltration in the final product. At last follow up 63 patients are alive and 10 died. The CXCR4 expression in CLL cells (n=73) and in CD34+ mobilized cells (n=32) was evaluated by multicolour flowcytometry analysis, CXCL12 G801A gene polymorphism by PCR-RFLP and SDF-1 plasma levels by ELISA Quantikine SDF-1alpha kit. There was an heterogeneous expression of CXCR4 in CLL cells at diagnosis and patients were clustered into 3 groups depending of ratio MFICXCR4 vs IgG1 isotype control: 37 (51%) patients with low CXCR4 rMFI (median=39 [range 8-86]), 23 (31%) patients with intermediate CXCR4 rMFI (median 148 [range 94-216]) and 13 (18%) patients with high CXCR4rMFI (mediane 286 [range 220-534]). Concerning CXCL12-3'A polymorphism study, 5 (7%) patients showed A/A, 16 (22%) G/A and 52(71%) G/G genotype. The median of SDF1 plasma level was 1520 pg/ml (1150.4-4907.2). We did not find any correlation neither between CXCR4 rMFI in CLL cells and CXCL12-3'A polymorphism (p=0.99) nor between CXCR4 rMFI and SDF1 plasma levels (p=0.73) but CXCR4 rMFI in CD34+ mobilized cells was correlated with CXCL12-3'A polymorphism (p=0.02). In addition, we didn't find any significant correlation between Binet stage, disease status after induction and CXCR4 rMFI in CLL cells. Among the 32 mobilized patients, the median of CXCR4 rMFI in CLL cells was 124 (11-534) and 19.5 (3-38) in CD34+ mobilized cells; the CXCL12-3'A polymorphism showed 2 A/A, 7 G/A and 23 G/G. We didn't find any correlation between CXCR4 rMFI in CLL cells and CXCR4 rMFI in CD34+ mobilized cells. Among the potential impacting factors on CD34+ cell mobilization, we didn't find any correlation with age (p=0.99), lymphocyte infiltration (p=0.36), CXCL12-3'A polymorphism (p=0.8) and plasma levels SDF1 (p=0.12). There was a significant impact of disease status (p=0.006) and a negative trend of CXCR4 rMFI (p=0.07).
In conclusion, the CXCR4 rMFI was very heterogeneous among our population. There was no correlation between this variable and CXCL12-3'A polymorphism. We showed a significant correlation between CD34+ mobilization and disease status, also a trend of CXCR4 rMFI. The CXCL12-3'AA genotype was associated with lower expression of CXCR4 in CD34+ mobilised cells and probably a good mobilisation; higher expression of CXCR4 in CLL cells at diagnosis was significantly correlated with poor CD34+mobilisation.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.