Introduction. In chronic lymphocytic leukemia (CLL), CD49d, the alpha chain of the heterodimer CD49d/CD29 (VLA-4), is a strong negative prognosticator, and a key player of tumor cell-microenvironment interactions. The adhesive properties of VLA-4 can be rapidly inside-out activated by signals through the B-cell receptor (BCR), thus favoring the capability of the integrin to interact with its specific ligands. In this context, VLA-4 needs to be maintained in an activated state by a continuous stream of inside-out signals from the BCR, which can be triggered by canonical antigens as well as in an autonomous antigen-independent manner, a BCR-mediated signaling recently emphasized to specifically occur in CLL cells.

Aim. To investigate the constitutive VLA-4 activation state in CLL cells and to connect this still not described feature with the presence of signals from the BCR continuously activating VLA-4.

Methods. Expression of the integrin alpha (CD49c, CD49d, CD49e) and beta1 (CD29) chains was analyzed by flow cytometry. The VLA-4 activation state was investigated by flow cytometry using the conformational sensitive anti-CD29 monoclonal antibody HUTS21, employed in conjunction with sources of VLA-4 ligands: either autologous plasma-containing fibronectin (FN) and soluble (s) VCAM-1 or increasing concentrations of exogenously added LDV-containing peptides as a VLA-4 specific ligand. In detail: i) HUTS21 expression was analyzed using whole blood (WB) samples from 727 CLL patients. Negative controls were obtained after plasma depletion through washing of WB samples; ii) the VLA-4 activation state expressed as receptor occupancy (RO) (J Biol Chem, 284,14337, 2009) was analyzed in sequential (t=0, day14, day 30) frozen/thawed samples from CLL patients treated in vivo with ibrutinib (IB) in real-world (n=16) and from a clinical trial (NCT02827617, n=14). BCR engagement was performed using goat F(ab′)2 anti-human IgM. ELISA assays were used to quantify FN and sVCAM-1 in plasma samples.

Results. According to the 30% cutoff, CD49d was expressed in 351/727 (48%), while CD29 was expressed in 676/727 (93%), and a strong correlation between CD49d and CD29 was observed (rho=0.7, p<0.0001). Among CD49d positive CLL cases, 81/351 (23%) expressed a constitutively activated form of VLA-4 identified by >20% (arbitrary cut-off) HUTS21 expression. Notably, HUTS21 positive expression was not circumscribed to VLA-4 expressing CLL, being also detected in 64/375 (17%) CD49d-, CD49c+ and/or CD49e+ CLL, thus envisioning a role of other integrins (CD49c and CD49e) in CLL cell interactions in tissues sites. HUTS21 expression did not correlate with a definite cytogenetic group, IGHV mutational status or with a specific IGHV family and stereotype. Depletion of the plasma component from the WB samples before HUTS21 staining significantly reduced the proportion of HUTS21 positive cells compared with those measured in WB samples (p=0.001, n=11), suggesting the need of plasma-borne ligands for HUTS21 epitope exposure. Consistently, both FN (mean 350 μg/ml) and sVCAM-1 (mean 4.8 μg/ml) were detected in the plasma of CLL cases, irrespective to HUTS21 positivity. According to these observations, variable constitutive VLA-4 activation states were observed in CLL cells collected at the pre-treatment stage from patients undergoing IB treatment (VLA-4 RO ranging from 0.22 to 0.40); these values significantly decreased in CLL cells collected at day 14 (p=0.03) and day 30 (p=0.02) on IB, suggesting an IB-dependent impairment of antigen-independent (autonomous) BCR signals. The variability of the constitutive VLA-4 activation levels observed at pre-treatment was paralleled by variable VLA-4 activation levels upon BCR triggering (VLA-4 RO ranging from 0.56 to 0.65). Of note, an inverse correlation between the VLA-4 constitutive level and the extent of anti-IgM induced-VLA-4 activation was found (n=19, rho=-0.5, p=0.0093), implying a competition between antigen-independent and antigen-driven BCR signalling.

Conclusion. The presence of a constitutively activated form of VLA-4 is observed in a subset of CLL which might be connected to continuous VLA-4 inside-out stimulation derived from an autonomous BCR signaling, which is currently under evaluation.

Disclosures

Zaja:Takeda: Honoraria; Sandoz: Honoraria; Amgen: Honoraria; Novartis: Honoraria, Research Funding; Celgene: Honoraria, Research Funding; Janssen: Honoraria; Abbvie: Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution