Abstract 2329

Poster Board II-306


In chronic lymphocytic leukemia (CLL), CD49d, often in association with CD38, has been shown to mark a disease subset with poor prognosis. Functionally, both molecules act as counter-receptors for surface structures (i.e. VCAM-1/CD106 and CD31) usually expressed by the endothelial/stromal component of tumor micro-environment. We have recently identified a micro-environmental circuitry which involves CD38 triggering, and eventually determines an enrichment of the VCAM-1/CD106-expressing endothelial component detected in the context of CLL infiltrates found in bone marrow biopsies. Data was also provided that CD49d/VCAM-1 interactions are active in delivering pro-survival signals to CD49d-expressing CLL cells (Zucchetto et al, Cancer Res, 69, 4001, 2009). In this study, we investigated the amount of circulating progenitors with endothelial phenotype in CLL samples with different CD49d and CD38 expression levels.


Peripheral blood (PB) samples from 91 CLL cases purposely selected with WBC>25,000/μl (B cells absolute lymphocyte count >10,000/μl) were evaluated by multiparametric flow cytometry for the absolute count of circulating CD34+ cells (ISHAGE protocol in single platform). Whenever possible (i.e. if a cluster of at least 100 CD34+ cells was detectable), a further characterization was performed (4-6 colours flow cytometry) for circulating endothelial cells (CEC), identified as a CD34+CD45low cell population co-expressing one of the following endothelial markers: CD309/VEGFR-2, CD144/VE-cadherin, CD106/VCAM-1 and CD146/Muc-18. CD49d and CD38 expression by CLL cells was considered positive if exceeding the standard cut-off value of 30% of positive cells.


PB absolute CD34+ cell counts were 7.5±7.5/μl in CD49d+ CLL (32 cases), vs. 3.3±2.7/μl in CD49d CLL (59 cases; p=2.6×10−4), or 9.4±8.7/μl in CD49d+ CLL (30 cases) vs. 4.6±2.9/μl in CD49d CLL (18 cases; p=0.004) when only cases phenotyped for CEC were considered. Furthermore, when samples were stratified also for CD38 expression, values of circulating CD34+ cells increased to 10.6±10.1/μl in CD38+CD49d+ CLL (11 cases) vs. 3.1±2.4/μl in CD38CD49d (51cases; p=1×10−5). Regarding the absolute quantification of CEC, a CD49d+ phenotype again marked the CLL subset with the highest CEC count, as identified by the expression of either the CD309/VEGFR-2 (CEC counts 1.7±2.3/μl in CD49d+ CLL vs. 0.5±0.5/μl CD49d CLL; p=0.009) or the CD144/VE-cadherin (CEC counts 0.8±1/μl in CD49d+ CLL vs. 0.3±0.5/μl in CD49d CLL; p=0.057) endothelial markers on CD34+CD45low cells. Notably, CEC from CD49d+ CLL expressed CD106/VCAM-1 in virtually all cells (1.6±2.4/μl), while the other marker of endothelial activation CD146/Muc-18 was detected in a fraction of CEC only (0.4±0.9/μl).


CD49d and CD38 expression by CLL cells identify a disease subset with significantly higher number of both circulating CD34+ cells and CEC. This phenomenon could be explained considering several aspects: i) the sharing of common phenotypic markers between CLL cells and CD34+ progenitors, including CD38 and CD49d, which could be responsible for a displacement of CD34+ progenitors in the context of micro-environmental niches; ii) the known capacity of CLL cells, especially with a unmutated IGHV gene status and/or a CD38+CD49d+ phenotype to produce pro-angiogenic factors including Ang-2; iii) the rare PB cells expressing CD34 and CEC markers may represent CLL cell precursors with tumor-initiating cell features. Studies are currently ongoing to dissect among these hypotheses


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.