CD20, CD22, CD23, but Not CD37 Expression on CD19⁺ B-Cells Is Altered by Exogenous Factors in a Sub-Population of Chronic Lymphocytic Leukemia/Small Lymphocytic Leukemia (CLL/SLL) Patients,

Chronic Lymphocytic Leukemia/Small Lymphocytic Leukemia (CLL/SLL) is a lymphoproliferative disorder that is characterized by the slow accumulation of malignant B cells. Patients follow heterogeneous clinical courses. The effectiveness of Rituximab, an anti-CD20 monoclonal antibody (mAb), is limited because target B cells from CLL patients express low levels of CD20. We previously reported that human serum suppresses CD20 expression ex vivo. Therefore, understanding mechanisms of low expression of CD20 as well as other target surface receptors would benefit CLL patients. New therapies such as epratuzumab (anti-CD22 mAb), lumiliximab (anti-CD23 mAb), and TRU-016 (anti-CD37 SMIP protein) are being investigated as alternative means to treat CLL. We therefore investigated CD20, CD22, CD23, and CD37 expression during ex vivo culture of CLL B cells from thirteen patients with and without human serum albumin (HSA), normal donor serum (NS), autologous CLL serum (CS), fetal bovine serum (FBS), IL2, IL4, IL13, IL15, IL21, TNFα, IFNα, IFNγ, G-CSF, or GM-CSF. Methods: Peripheral blood (PB) from thirteen CLL patients and five healthy normal donors was obtained with IRB approval. Nine CLL patients were Rai 0/1, one Rai 4, and three unknown. Prognosis, as determined by cytogenetics and/or FISH, was: four good, three intermediate, one poor, and five unknown. Beta-2 Microglobulin (B2M) ranged from 1.4–7.1 mg/L. CD19⁺ cells, isolated by positive magnetic selection, were cultured in AIM-V serum-free media at 37°C, 5% CO₂, 100% humidity for 0, 24, 48, and 96 hours. Cells were either untreated, treated with 5% serum, or 5ng/mL cytokine. Receptor expression was measured by multivariate flow cytometry and calculated as percent loss or increase in response to treatment. Data was presented as mean ± SEM and analyzed using the Mann-Whitney U test as compared to HSA treatment. Results: Prior to culture, CD19⁺ CLL cells had 11.7±1.7% CD20, 40.3±7.9% CD22, 60.6±6.8% CD23, and 81.6±2.7% CD37 positive expression (N=13). Analysis of untreated CLL cells cultured for 96 hours showed that the CLL patients subdivided into two groups based on changes in CD20 expression: variable (72.5±7.4% CD20⁺, N=9) and stable (27.2±4.2% CD20⁺, N=4). Within the variable group, losses in CD20 expression resulted from treatment as compared to HSA were: NS (40.2±3.6%, p<0.05), CS (43.6±8.8%, p<0.05), FBS (37.2±5.0%, p=0.064), IL2 (11.5±5.1%, p=0.860), IL4 (33.3±8.9%, p=0.185), and IL13 (24.2±9.1%, p=0.596). The percent loss in CD20 expression due to IL4 was significantly greater than IL2 (p<0.05). Percent losses in CD22 expression were: NS (62.4±9.4%, p<0.05), CS (63.2±8.1%, p<0.05), and FBS (66.6±6.8%, p<0.05). No loss in CD37 expression was significant. Percent increases in CD23 expression were: IL4 (253.7±79.4%, p=0.930) and IL13 (86.7±77.9%, p=0.377). Within the stable group after treatment, percent losses in CD22 were: IL2 (40.0±10.1%, p=1.000) and IL4 (29.1±6.5%, p=0.114). Percent increase in CD37 expression was: FBS (53.3±18.3%, p<0.05). Conclusions: Our findings show that cells from CLL patients form groups based on the ability of cells to maintain low CD20 expression after serum-free culture for 96 hours: variable and stable. Only CLL cells within the variable group have a high upregulation of CD20. These patient samples respond to serum through a reduction of CD20. In addition, suppression of CD20 expression due to IL4 or IL13 treatment was similar to that of serum. Both IL4 and IL13 share a common receptor that signals through the JAK/STAT pathway, suggesting downstream signals may be responsible for suppression of CD20 expression in CLL. Similarly, CLL patients belonging to the variable group also have reduced expression of CD22 in response to serum. Loss of CD23 expression in CLL cells cultured over time, in the absence of IL4 or IL13 treatment, may be predictive of poor clinical outcomes in patients treated with the anti-CD23 monoclonal antibody. CD20, CD22, and CD23 are susceptible to environmental cues, whereas expression of CD37 remained high throughout culture and independent of treatment. As a result, we predict CD37 to be a good target for treating CLL due to its stability. Ongoing clinical trials with TRU-016 or other agents that specifically target CD37 may therefore be of potential therapeutic benefit.

No relevant conflicts of interest to declare.