Mechanisms of the BCR-Mediated CXCR4 Down-Regulation and Its Clinical Relevance in Chronic Lymphocytic Leukemia Progression

Antigen-driven signals are involved in the pathogenesis and progression of CLL with a particular occurrence within the lymph node microenvironment. We have previously shown that ex-vivo BCR engagement promoted simultaneous down-regulation of CXCR4 and CD62L membrane expressions in CLL cells from patients at risk of disease progression only. Functionally, cells down-regulating CXCR4 and CD62L in response to BCR triggering displayed both a reduction in migration toward CXCL12 and in adhesion to lymphatic endothelial cells.

We studied a cohort of 73 previously untreated patients, 36 of whom were IGHV mutated and 37 were IGHV unmutated. The distribution of the extent of CXCR4 down-regulation was identical among CXCR4 responders, whether IGHV mutated (median 49% [Q1:28-Q3:62]) or unmutated (median 43% [Q1:31-Q3:66]). We also studied the expression of another chemokine receptor, CXCR5, which is also expressed at the membrane of CLL cells and plays an important role in the homing and trafficking of lymphocytes to the lymphoid follicles. Sustained antigenic stimulation of CLL cell samples (n=25) promoted down-regulation of CXCR4 and CXCR5 in a strictly correlated manner (r= 0,9) i.e. in the same subsets of cells. The extent of BCR-dependent decrease of the two proteins was strikingly representative of the heterogeneous capacity of the B-CLL cells to respond to BCR engagement in a given patient.

Moreover, treatment with the specific PKD inhibitor CID755673 blocked significantly both CXCR4 and CXCR5 BCR-mediated decrease (n=11), demonstrating that PKDs specifically target these 2 molecules, whereas CD62L down-regulation was not significantly affected by the PKD inhibitor. At the functional level, treatment with the PKD inhibitor restored CLL cell migration capacity in response to CXCL12. PKD phosphorylation/activation in response to BCR stimulation, which involves PI3K-d, was required for CXCR4-phosphorylation and its down-regulation.

We then studied the clinical relevance of CXCR4 down-regulation after BCR engagement. The capacity to down-regulate CXCR4 was significantly related to shorter PFS (p=0.043). This cohort allowed exploring the link with IGHV mutational status. 36/37 unmutated IGHV cases had a significant CXCR4 down-regulation after in vitro BCR stimulation, in concordance with the ultimately constant progression of the disease. Conversely, IGHV mutated cases fell into two groups: a) one group of 14/36 patients, with very low (<5%) or no CXCR4 decrease in response to BCR ligation. None of these patients had any lymph node enlargement with a median follow up of 8.4 years [Q1:4.4-Q3:9.9]. Four patients among them required treatment but all for other reasons than nodal progression. b) the other group of 22/36 patients, for which there was a variable but significant CXCR4 down-regulation after BCR engagement. Among these cases, with a median follow up of 9.2 years [Q1:6.4-Q3:10.7], 14 patients developed clinical lymphadenopathy and 7 patients required treatment, all of them for tumor progression. The difference in lymphadenopathy was highly significant between these 2 IGHV mutated groups (p< 0.001).

In conclusion, the capacity of cells to down-regulate cell surface chemokine receptors and L-selectin is reflecting the size of the cell subset responding to BCR signaling. Interestingly, BCR signaling capacity to down-regulate CXCR4 is variable among IGHV mutated samples and strongly linked to the presence of lymph nodes as well as disease progression with time. It is tempting to speculate that depending on the BCR-mediated CXCR4/CXCR5 down-regulation capacity, some IGHV mutated patients might not benefit from Btk or PI3K-d inhibitors. Given the crucial importance of down-regulation of these two chemokine receptors after BCR engagement, we suggest that PKD could be promising new therapeutic target in CLL.