Abstract 2326

Poster Board II-303

B-cell type chronic lymphocytic leukemia (B-CLL), an incurable disease of unknown etiology, results from the clonal expansion of a CD5+CD19+ B lymphocyte. Progress into defining the cell of origin of the disease and identifying a stem cell reservoir has been impeded because of the lack of a reproducible model for growing B-CLL cells in vivo. At least one possible cause for this is the murine microenvironment's inability to support B-CLL survival and proliferation. To overcome this barrier, we reconstituted NOD/SCID/γnull mice by intrabone (ib) or intravenous (iv) injection of 1 × 105 CD34+ cord blood cells along with ∼106 bone marrow-derived human mesenchymal stem cells (hMSCs) by ib injection. After human cellular engraftment, a total of 108 CFSE-labeled PBMCs from individual B-CLL patients were injected into the same bones or iv. Every two weeks thereafter, blood from the mice was examined for the presence of cells bearing CFSE, human CD45, and various human lineage markers by flow cytometry. In the presence of a human hematopoietic microenvironment derived from hHSCs, CFSE+CD5+CD19+ cells were readily detected in the blood of mice and many of these leukemic cells underwent at least 6 cell doublings. In contrast, the numbers of leukemic cells circulating in the blood of mice reconstituted with hMSCs without CD34+ cells was much less and these failed to proliferate. Thus, hMSCs were not essential in the model. Moreover, the percentage of leukemic cells expressing CD38 in the CFSE+CD5+CD19+ cell fraction was similar to that in the donor patient inoculum only in the mice in which B-CLL cell proliferation occurred. The percentage and intensity of CD38-expressing B-CLL cells was higher in the spleen and bone marrow (BM), far exceeding that in the blood and peritoneum. Of note, B-CLL cells formed follicular structures in the spleen that contained larger B cells expressing the same Ig H and Ig L chains as the CLL MNCs. CLL cells from these spleens exhibited the same IGHV/D/J rearrangement as in the donor leukemic cells, indicating their leukemic origin. These follicular structures are reminiscent of proliferation centers/pseudofollicles seen in patient lymph nodes and BM, in that they contained leukemic B cells of intermediate and large size. Finally, B-CLL cells adoptively transferred into these mice exhibit kinetics similar to those observed in patients in vivo, with birth rates calculated from CFSE-dilution data of (X-Y% per day). Robust T-cell expansion occurred in mice receiving CD34+ cells and occasionally in mice (10-20%) not receiving hCD34+ cells. Based on detailed SNP analyses, the expanded T cells were of B-CLL patient origin and not from hCD34+ cells. Notably when T cells were eliminated by injecting an anti-CD3 mAb (OKT3), B-CLL cell proliferation was inhibited. These latter two findings clearly indicate a need for autologous T lymphocytes in the successful adoptive transfer of B-CLL cells in this model and also highlight the fact that another cell type, derived from the normal allogeneic CD34+ cells, is needed. Because CD34+ cells could be substituted for by mature allogeneic monocytes or B lymphocytes from the blood of normal individuals, antigen-presenting cells may be the key cells that develop from the normal hHSCs. This would suggest that an allogeneic mixed lymphocyte reaction is needed for successful B-CLL cell survival and proliferation. This is consistent with finding that most of the mice with significant T-cell overexpansion died within 6 weeks of B-CLL cell injection from apparent graft vs. host disease. Finally, preliminary data using Rituxan (anti-CD20 mAb) to eliminate B-CLL cells from recipient mice suggest that this model is a good tool to perform pre-clinical studies on the efficacy of action of novel therapeutics. In summary, these studies indicate that allogeneic human antigen-presenting cells and autologous human T cells permit adoptive xenogeneic transfer and clonal expansion of B-CLL cells in immune deficient mice. This model will be useful in discovering and understanding non-genetic factors promoting B-CLL expansion because it recapitulates several features of human B-CLL. Finally, the model may help in the study of the basic biology of this disease, such as if leukemic stem cells exist, and also in conducting pre-clinical tests on possible new therapeutics.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.