Variables affecting the quantitation of CD22 in neoplastic B cells

Abstract

Background: Quantitative flow cytometry (QFCM) is being applied in the clinical flow cytometry laboratory for diagnosis, prognosis, and assessment of patients receiving antibody-based therapy. ABC values and the effect of technical variables on CD22 quantitation in acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), mantle cell lymphoma (MCL), follicular lymphoma (FCL), hairy cell leukemia (HCL) and normal B cells were studied.

Methods: The QuantiBrite System® was used to determine the level of CD22 expression (mean antibody bound per cell, ABC) by malignant and normal B cells. The intra-assay variability, number of cells required for precision, effect of delayed processing as well as shipment of peripheral blood specimens (delayed processing and exposure to noncontrolled environments), and the effect of paraformaldehyde fixation on assay results were studied.

Results: The QuantiBRITE method of measuring CD22 ABC is precise (median CV 1.6%, 95% confidence interval, 1.2-2.3%) but a threshold of 250 malignant cells is required for reliable CD22 ABC values. Delayed processing and overnight shipment of specimens resulted in significantly different ABC values whereas fixation for up to 12 h had no significant effect. ABC measurements determined that CD22 expression is lower than normal in ALL, CLL, FCL, and MCL but higher than normal in HCL.

Conclusions: CD22 expression was atypical in the hematolymphoid malignancies studied and may have diagnostic utility. Technical variables such as cell number analyzed and delayed processing or overnight shipment of specimens impact significantly on the measurement of antigen expression by QFCM in the clinical laboratory.

Fig. 1.

Analysis gating strategies. A. Chronic lymphocytic leukemia (CLL): (1) Ungated data—Gate drawn on CD19 positive B cells. (2) CD19 positive B cells—Gate drawn on CD19 and CD5 positive CLL cells. (3) CD 19 and CD5 positive CLL cells: CD22 expression. FITC channel is blank. B. Hairy cell leukemia (HCL): (1) Jngated data—Gate drawn on CD20 positive B cells. (2) CD20 positive B cells—Gate drawn on CD20 and CDllc bright positive HCL cells. (3) CD20 and CDllc positive HCL cells: CD22 expression. FITC channel is blank. C. Acute lymphoblastic leukemia (ALL): (1) Ungated data—Gate drawn on CD19 positive cells. (2) CD19 positive cells—Gate drawn on CD19 and CD10 positive ALL cells. (3) CD19 and CD10 positive ALL cells: CD22 expression.

Fig. 2.

Optimal cell number for reliable CD22 ABC values. Twenty eight samples were analyzed multiple times collecting 10, 20, 30, 40, 50, 100, 250, 500, 1,000, 2,500, and 5,000 malignant/aberrant cells. The standard deviation (SD) of the relative deviations of CD22 ABC from the sample average (y axis) was plotted versus number of aberrant cells the CD22 ABC was determined from.