CD19 has been as a classic marker for gating B-cells in flow cytometric analysis. However, with the application of CD19 chimeric antigen receptor (CAR) T-cell therapy for treating relapsed/refractory B-cell acute lymphoblastic leukemia (B-ALL), CD19 gating is no longer adequate for B-ALL patients post CD19 CAR-T therapy owing to CD19 antigen loss, under this circumstance, cytoplasmic CD79a (cCD79a) emerged to replace CD19 as a gating marker of B-cells. Nevertheless, cCD79a was intracellular staining and time consuming, sometimes with unspecific stain or without enough cells for analysis after permeabilization and extra washing. CD22 is another pan B marker expressed on the surface of B lymphocytes at different stages, could be as alternative choice of B-cell identification after CD19 CAR-T. Here, we evaluated the possibility of CD22 as a B-cell gating marker by comparing the expression of CD22 and cCD79a in CD19 negative relapsed B-ALL patients.

We routinely used cCD79a to gate B-cells in patients after CD19 CAR-T therapy, CD22 expression was detected simultaneously. The standard antibody panel consist of CD45, CD19, CD22, CD10, CD20, CD34, CD38, CD58, cCD79a, CD81, HLA-DR and CD123. CD22 antibody conjugated with phycoerythrin (PE) or allophycocyanin (APC) was used, extra antibodies could be added based on patient's immunophenotype at initial diagnosis. Bone marrow specimens were prepared by a standard stain-lyse-wash procedure, no less than 300,000 events were acquired by 8-color BD FACS Canto II flow cytometer, data were analyzed using FACS Diva 8.0.1 software. Antigen normal expression and partial expression were defined as >80% and 20-80% of gated cells displaying interested antigen, respectively. Dim expression was defined as more than one log weaker than the normal counterpart's mean fluorescence intensity (MFI).

From April 2018 to June 2020, a total of 40 CD19-negative relapsed B-ALL patients post CD19 CART therapy were included in this study, consisting of 14 adults and 26 children younger than 18 years old, with a median age of 12 years (range, 1-64). The simultaneous expression of cCD79a and CD22 among these patients were analyzed. All 40 patients (100%) had normal cCD79a expression on B-cells. Although CD22 was expressed on all cases as well, normal expression was seen in 33 (82.5%) patients whereas 7 (17.5%) were partial or dim (P/D) expression, the P/D expression rendered CD22 as a gating marker unsuitable. To figure out whether this P/D CD22 expression was related to CD19 CAR-T treatment, we checked back the immunophenotype of these 7 patients and found that, before CD19 CAR-T, 4 (10%) patients with P/D CD22 expression and 3 (7.5%) with normal CD22 expression. When we excluded these 4 patients with existed P/D CD22 expression, 91.7% (33/36) of patients accordantly expressed normal CD22 and cCD79a, which indicated that CD22 could be a B-cell gating marker for 91.7% of patients with normal CD22 expression before CD19 CAR-T therapy.

To obtain optimal signal, CD22 antibody conjugated with bright fluorescein such as PE or APC is recommended. It is noted that CD22 also expressed on basophils, therefore, HLA-DR should be included in the antibody panel to exclude this interference since basophils do not express HLA-DR. Sometimes, there might be CD22 stain on plasmacytoid dendritic cells (pDCs) which express HLA-DR too, it is not difficult for us to differentiate B-cell from pDCs based on the facts that CD22 expression on pDCs is weaker than that on B cells, pDCs have bright CD123 expression and are located at a relatively fixed position in CD45/SSC plot.

In conclusion, our data on the comparison between CD22 and cCD79a expression in CD19 negative relapsed B-ALL patients revealed that, under most circumstances (with 82.5%-91.7% chance), CD22 gating alone could efficiently identify B cells in patients after CD19 CAR-T therapy, which is also a cost-effective and labor-saving strategy for routine practice compared to cCD79a gating. In case of partial or dim CD22 expression, the cCD79a gating is needed.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.