Partial Characterization and In Vitro Expansion of Putative CLL Precursor/Stem Cells Which Are Dependent on Bone Marrow Microenvironment for Survival

Abstract 2433 Chronic lymphocytic leukemia (CLL) is characterized by the clonal expansion of B lymphocytes which typically express CD19 and CD5. The disease remains incurable and recurrence often occurs after current standard therapies due to residual disease or probably due to the presence of therapy-resistant CLL precursors. Based on the growing evidence for the existence of leukemia stem cells, this study was designed to search for putative CLL precursors/stem cells based on the co-expression of CLL cell markers (CD19/CD5) with the hematopoietic stem cell marker (CD34). Forty seven CLL patients and 17 healthy persons were enrolled in the study. Twenty four patients had no previous treatment and 23 had pre-therapy. Twenty two patients were in Binet stage C and 25 patients in B. Twenty two patients had unmutated and 18 mutated IgVH gene (7: ND). Cytogenetic analysis by FISH showed that 14 patients had del 13q, 8 had del 11q, 4 had del 17p and 9 had trisomy 12. Peripheral blood and bone marrow mononuclear cells were subjected to multi-colour FACS analysis using anti-human antibodies against CD34, CD19 and CD5 surface antigens. The results revealed the presence of triple positive CD34+/CD19+/CD5+ cells in CLL samples (mean 0.13%; range 0.01–0.41) and in healthy donors (0.31%; range 0.02–0.6) within the CD19+ B cells. However, due to the high leukocyte count in CLL patients, the absolute number of these cells was significantly higher in CLL samples (mean: 78.7; range 2.5–295 cells /μL blood) compared to healthy persons (mean: 0.45: range 0.04–2.5 cells/μl)(p<0,001). These triple positive “putative CLL stem cells” (PCLLSC) co-express CD133 (67%), CD38 (87%), CD127 (52%), CD10 (49%), CD20 (61%), CD23 (96%), CD44 (98%) and CD49d (74%). FISH analysis on 4 patients with documented chromosomal abnormalities detected the corresponding chromosomal aberrations of the mature clone in the sorted CD34+/CD5+/CD19+ and/or CD34+/CD19-/CD5- cells but not in the CD3+ T cells. Multiplex RT-PCR analysis using IgVH family specific primer sets confirmed the clonality of these cells. Morphologically, PCLLSC appeared larger than lymphocytes with narrow cytoplasm and showed polarity and motility in co-culture with human bone marrow stromal cells. Using our co-culture microenvironment model (Shehata et al, Blood 2010), sorted cell fractions (A: CD34+/19+/5+, B: CD34+/19-/5- or C: CD34-/CD19+/5+) from 4 patients were co-cultured with primary autologous human stromal cells. PCLLSC could be expanded in the co-culture to more than 90% purity from fraction A and B but not from fraction C. These cells remained in close contact or migrated through the stromal cells. PCLLSC required the contact with stromal cells for survival and died within 1–3 days in suspension culture suggesting their dependence on bone marrow microenvironment or stem cell niches. RT-PCR demonstrated that these cells belong to the established CLL clone. They also eexpress Pax5, IL-7R, Notch1, Notch2 and PTEN mRNA which are known to play a key role in the early stages of B cells development and might be relevant to the early development of the malignant clone in CLL. Using NOD/SCID/IL2R-gamma-null (NOG) xenogeneic mouse system we co-transplanted CLL cells from 3 patients (5 million PBMC/mouse) together with autologous bone marrow stromal cells (Ratio: 10:1). The percentage of PCLLSC in the transplanted PBMC was 0.18% (range 0.06–0.34%). Using human-specific antibodies, human CD45+ cells were detected in peripharal blood of the mice (mean 0.9 % range 0.47–1.63%) after 2 months of transplantation. More than 90% of the human cells were positive for CD45 and CD5. Among this population, 26% (range 15–35%) of the cells co-expressed CD45, CD19, CD5 and CD34 and thus correspond to the PCLLSC. In conclusion, our data suggest the existence of putative CLL precursors/stem cells which reside within the CD34+ hematopoietic stem cell compartment and carry the chromosomal aberrations of the established CLL clone. These cells could be expanded in vitro in a bone marrow stroma-dependent manner and could be engrafted and significantly enriched in vivo in NOG xenotransplant system. Further characterization and selective targeting and eradication of these cells may pave the way for designing curative therapeutic strategies for CLL. Disclosures: No relevant conflicts of interest to declare.