Comparison of the phenotype of Tfh subpopulations in whole-blood lymphocytes from alloimmunized patients, nonalloimmunized patients, and HDs. (A) Gating strategy for flow cytometry analysis. Lymphocytes were gated on size (forward scatter [FSC]) and granularity (side scatter [SSC]). Dead cells were excluded with Aqua LIVE/DEAD staining. Analyses were performed on CD4⁺CD45RA⁻ cells. The Tfh population was identified with the CXCR5⁺PD1⁺ gate. The Tfh PD1⁺⁺⁺ and Tfh PD1⁺ populations were identified with the CD4⁺CD45RA⁻CXCR5⁺ICOS⁺PD1⁺⁺⁺ and CD4⁺CD45RA⁻CXCR5⁺ICOS⁺PD1⁺ gates, respectively. The Tfh1, Tfh2, and Tfh17 populations were identified with the CD4⁺CD45RA⁻CXCR5⁺CCR6⁻CXCR3⁺, CD4⁺CD45RA⁻CXCR5⁺CCR6⁻CXCR3⁻, and CD4⁺CD45RA⁻CXCR5⁺CCR6⁺CXCR3⁻ gates, respectively. (B) Comparison of Tfh percentages in whole blood between alloimmunized (n = 14, black circles, 14 experiments, with 1 donor per experiment) and nonalloimmunized (n = 9, black squares, 9 experiments, with 1 donor per experiment) transfused SCD patients. Ethnically matched HDs (n = 10, black triangles, 10 experiments, with 1 donor per experiment) were used as a control group. (C) Comparison of PD1 expression on the Tfh CXCR5⁺ICOS⁺ subpopulation in whole-blood lymphocytes from patients and HDs. (D) Comparison of CXCR3 and CCR6 expression between Tfh subsets in whole-blood lymphocytes from patients and HDs. (E) Comparison of the PD1⁺ and PD1⁺⁺⁺ subsets in the Tfh1, Tfh2, and Tfh17 subpopulations in whole-blood lymphocytes from patients and HDs. Horizontal bars indicate the median values. Significant P values (< .05) were obtained in analysis of variance and post hoc tests. *P < .05; **P < .01; ***P < .005; ****P < .001.