Alterations in B- and circulating T-follicular helper cell subsets in immune thrombotic thrombocytopenic purpura

Abstract

T follicular helper (Tfh) cells regulate development of antigen-specific B-cell immunity. We prospectively investigated B-cell and circulating Tfh (cTfh) cell subsets in 45 patients with immune thrombotic thrombocytopenic purpura (iTTP) at presentation and longitudinally after rituximab (RTX). B-cell phenotype was altered at acute iTTP presentation with decreased transitional cells and post-germinal center (post-GC) memory B cells and increased plasmablasts compared with healthy controls. A higher percentage of plasmablasts was associated with higher anti-ADAMTS13 IgG and lower ADAMTS13 antigen levels. In asymptomatic patients with ADAMTS13 relapse, there were increased naïve B cells and a global decrease in memory subsets, with a trend to increased plasmablasts. Total circulating Tfh (CD4+CXCR5+) and PD1+ Tfh cells were decreased at iTTP presentation. CD80 expression was decreased on IgD+ memory cells and double-negative memory cells in acute iTTP. At repopulation after B-cell depletion in de novo iTTP, post-GC and double-negative memory B cells were reduced compared with pre-RTX. RTX did not cause alteration in cTfh cell frequency. The subsequent kinetics of naïve, transitional, memory B cells and plasmablasts did not differ significantly between patients who went on to relapse vs those who remained in remission. In summary, acute iTTP is characterized by dysregulation of B- and cTfh cell homeostasis with depletion of post-GC memory cells and cTfh cells and increased plasmablasts. Changes in CD80 expression on B cells further suggest altered interactions with T cells.

Graphical abstract

Figure 1.

B cell subsets at iTTP presentation and ADAMTS13 relapse. Percentages of transitional cells (IgD⁺CD38⁺⁺) (A), naïve cells (IgD⁺CD38⁺) (B), IgD⁺ memory B cells (IgD⁺CD38^–) (C), postgerminal center memory B cells (IgD^–CD38⁺) (D), double–negative memory B cells (IgD^–CD38^–) (E), and plasmablasts (IgD^–CD38⁺⁺) (F) in patients with acute iTTP episodes (n = 22), ADAMTS13 relapse (n = 24), and HC (n = 27). A13 relapse, ADAMTS13 relapse; double-negative memory cells, double-negative memory cells; Post GC memory, postgerminal center memory cells.

Figure 2.

Circulating Tfh subsets at iTTP presentation and ADAMTS13 relapse. Percentages of total cTfh (CD4⁺CXCR5⁺) (A), PD1⁺ cTfh (CD4⁺CXCR5⁺PD1⁺) (B), ICOS⁺ cTfh (CD4⁺CXCR5⁺ICOS⁺) (C), and PD1⁺ICOS⁺ cTfh (CD4⁺CXCR5⁺PD1⁺ICOS⁺) (D) in patients with acute iTTP episodes (n = 34), ADAMTS13 relapse (n = 27), and HC (n = 27).

Figure 3.

Relationship between ADAMTS13 antigen levels and plasmablast frequency. Spearman correlation analysis was performed, and P < .05 indicates that the difference is statistically significant.

Figure 4.

Analysis of CD80 expression (median fluorescence intensity) on B-cell subsets defined by IgD/CD38. Acute iTTP cases (A) and ADAMTS13 relapse cases (B).

Figure 5.

Pairwise comparison of B-cell subset frequencies before and after RTX by Wilcoxon signed–rank test. (A) Acute iTTP episodes, pairwise comparison of 10 patients. (B) ADAMTS13 relapse episodes, pairwise comparison of 13 patients.

Figure 6.

Longitudinal changes in B-cell subsets after RTX therapy for an acute iTTP episode or ADAMTS13 relapse (ADAMTS13 activity 15%) until subsequent ADAMTS13 relapse, compared with patients who remained in remission over an equivalent period.