Comparative Assessment of Surface CD19 and CD20 Expression on B-Cell Lymphomas from Clinical Biopsies: Implications for Targeted Therapies

Background: CD19 and CD20 are B-cell lineage-specific antigens expressed on the cell surface of most B-cell lymphomas. While CD20 is acquired during late stages of B-cell lymphogenesis and is then lost upon differentiation into plasma cells, CD19 expression covers the entire spectrum of early B-cell genesis and maturation. CD20-targeting agents have been broadly integrated into the therapeutic armamentarium for B-cell lymphomas. More recently, CD19-targeting agents have emerged as promising alternatives with demonstrated therapeutic value. Given the imminent availability of both CD19 and CD20 targeted therapies, and potential for combinational approaches, we studied the surface expression of these antigens at the single-cell level on lymphoma cells and benign background lymphoid subsets from biopsy specimens.

Methods: Flow cytometric analysis (seven-color) was performed on biopsy specimens from 47 patients with newly diagnosed B-cell lymphomas, including diffuse large B-cell lymphoma (n=15), follicular lymphoma (n=15), marginal zone lymphoma (n=9), mantle cell lymphoma (n=9), Burkitt lymphoma (n=2), and unclassifiable low-grade B-cell lymphoma (n=2). Small lymphocytic lymphoma was intentionally excluded, given its well-described loss of CD20 expression. Biopsies from eight additional patients with persistent/recurrent B-cell lymphomas after anti-CD20 therapy (greater than 6 months after last dose) were also evaluated. Thresholds for CD19 or CD20 antigen positivity were defined for each case, based on the 95th percentile fluorescence intensity of the respective marker on tumor-infiltrating T-cells (internal negative control). In addition, CD19 or CD20 fluorescence intensities of tumor cells were normalized to background benign B-cells (internal positive control) using the median fluorescence ratio (MFR).

Results: Both CD19 and CD20 were highly expressed on CD20 treatment naïve tumor cells, with a slightly higher median percentage of positive tumor cells for CD19 (98%) compared with CD20 (93%) (p=0.003), and one case lacking CD20 expression. When compared with background benign B-cells, CD20 was frequently overexpressed on tumor cells (mean MFR=1.8), while CD19 expression was overall similar to background benign B-cells (mean MFR=0.9) (p=0.001). As the surface density of CD20 on benign B-cells is reportedly higher than CD19 (~100,000 vs ~20,000 molecules per cell, respectively), these findings are consistent with a higher density of surface CD20 than CD19 on most B-cell lymphomas. However, CD20 expression was more heterogeneous (within individual patient samples and across patients), with a higher median percentage of CD20-negative tumor events (median=0.5%, range=0-98%) compared with CD19-negative events (median=0%, range=0-28%) (p=0.003). Interestingly, expression of CD19 within the CD20-negative tumor subsets was largely preserved (mean % CD19-positive events=97.86%, min=40%). In addition, percentages of CD19/CD20 double-negative tumor events were very small (median=0%, range=0-4.9%), and only detectable in 15 cases (32%). Of eight additional cases studied post-anti-CD20 immunotherapy (6-84 months after last dose), the percentage of antigen-positive events by tumor cells was largely preserved for both CD19 (median=99.6%, range=93.5-100%) and CD20 (median=95.7%, range=55.6-100%), similar to the pre-therapy cohort.

Conclusions: CD19 and CD20 are both highly and consistently expressed in B-cell lymphomas. While CD20 has a higher average density of surface molecules per tumor cell, CD19 expression is more homogenous and is preserved in small CD20-negative tumor subsets and after anti-CD20 targeted therapy. These findings support the clinical evaluation of anti-CD19 immunotherapies and combinational therapies targeting both surface antigens.