T cell Activation Marker HLA-DR Reflects Tacrolimus-Associated Immunosuppressive Burden and BK Viremia Risk After Kidney Transplantation - An Observational Cohort Study

Abstract

Kidney transplantation (KT) is the current treatment of choice in patients with end-stage kidney disease. Immunosuppression is required to prevent acute rejection but is associated with a high incidence of adverse events. The immunosuppressive burden substantially differs between individuals, necessitating new immune monitoring strategies to achieve personalization of immunosuppression. To compare the evolution of T cell profiles in correlation with immunosuppression and clinical outcomes, 87 kidney transplant recipients were followed for 12 months after KT. Flow cytometry along with assessment of T cell activation markers and clinical data was performed before KT and during study visits 10 days, 2 months and 12 months after KT. Longitudinal T cell phenotyping revealed a significant decrease of T cell activation markers HLA-DR, FCRL3, and CD147 in CD4⁺ effector T cells after KT. The most pronounced reduction (75%) was found for the activation-proliferation marker HLA-DR, which persisted throughout the observational period. The decrease in HLA-DR expression reflected immunosuppressive burden through strong associations with tacrolimus trough-level exposure (coeff = -0.39, p < 0.01) and BK viremia incidence (coeff = -0.40, p < 0.01) in multivariable regression analysis. T cell activation marker HLA-DR emerges as a potential biomarker for tacrolimus-related immunosuppressive burden in association with BK viremia risk following KT.

Keywords: immune monitoring; immunosuppression; kidney transplantation; personalized medicine; translational nephrology.

FIGURE 1

T cell activation marker HLA-DR identifies a CD4⁺ T cell subset susceptible to immunosuppression after KT. (A) CD3⁺CD4⁺CD25⁻CD127⁺ T_eff were clustered by activation status using FlowSOM algorithm and ClusterExplorer in FlowJo analysis software from peripheral PBMCs isolated immediately before and 2 months after KT. (B, C) The longitudinal evolution of absolute cell counts and frequencies are shown as box blots (MDN ± IQR) for all study visits with multiple group comparison by mixed-effects analysis; significant results are shown by asterisks (**) p < 0.01, (***) p < 0.001, (****) p < 0.0001. (D) Representative raw flow cytometry contour plots of one patient for each timepoint.

FIGURE 2

Effector T_reg replenish after induction therapy. (A) CD3⁺CD4⁺Foxp3⁺CD127^- T_reg were clustered by activation status using FlowSOM algorithm and ClusterExplorer in FlowJo analysis software from peripheral PBMCs isolated immediately before (preKT) and 2 months after KT (M2). Temporary decrease of absolute counts and frequencies of (B, C): activated CD45RA⁻CD15s⁺ T_reg and (D, E): Ki67⁺ proliferative-effector T_reg after KT; box blots (MDN ± IQR) for all study visits with multiple group comparison by mixed-effects analysis; significant results are shown by asterisks (***) p < 0.0001, (****) p < 0.00001.

FIGURE 3

HLA-DR⁺ T_eff counts strongly correlate with tacrolimus trough level exposure. (A) Median tacrolimus trough level (TL) trend over time is shown as a red line. The area under the curve (AUC) was calculated by the trapezoidal rule (median AUC = 113.7 ng*t/mL) to represent tacrolimus TL exposure. TL exposure (TL AUC) was then plotted against the (B): mean HLA-DR⁺ T_eff count and (C): proliferative-effector Treg counts of individual patients starting at D10 until M2; Spearman correlation coefficient (r) was calculated to determine the strength of the relation.

FIGURE 4

HLA-DR⁺ T_eff counts are significantly lower in patients developing BKV viremia. (A) The mean HLA-DR⁺ T_eff counts and (B): the mean proliferative-effector Treg counts between D10 and M2 of individual patients were pairwise compared between event and no event groups for BKV, CMV, and BPAR. (**) indicates p < 0.01.

FIGURE 5

T cell activation marker HLA-DR is associated with BK viremia, potentially allowing risk stratification early after KT. (A) Predicted probabilities for the first incident of BK viremia from cox regression stratified by HLA-DR⁺ T_eff counts are depicted with a best-fit line (blue line); aHR = 1.49 [1.24–1.80] per unit decrease of HLA-DR, p = 0.00002. (B) Patients were stratified by HLA-DR⁺ T_eff count at day 10 above or below 4.71 × 10³/mL (cutoff determined by tdROC analysis of viremia incidence with AUC of 0.75; p = 0.002) to display the risk difference for experiencing BK viremia by Kaplan-Meier curves with log-rank analysis.