Effect of Optimized Immunosuppression (Including Rituximab) on Anti-Donor Alloresponses in Patients With Chronically Rejecting Renal Allografts

Abstract

RituxiCAN-C4 combined an open-labeled randomized controlled trial (RCT) in 7 UK centers to assess whether rituximab could stabilize kidney function in patients with chronic rejection, with an exploratory analysis of how B cell-depletion influenced T cell anti-donor responses relative to outcome. Between January 2007 and March 2015, 59 recruits were enrolled after screening, 23 of whom consented to the embedded RCT. Recruitment was halted when in a pre-specified per protocol interim analysis, the RCT was discovered to be significantly underpowered. This report therefore focuses on the exploratory analysis, in which we confirmed that when B cells promoted CD4+ anti-donor IFNγ production assessed by ELISPOT, this associated with inferior clinical outcome; these patterns were inhibited by optimized immunosuppression but not rituximab. B cell suppression of IFNγ production, which associated with number of transitional B cells and correlated with slower declines in kidney function was abolished by rituximab, which depleted transitional B cells for prolonged periods. We conclude that in this patient population, optimized immunosuppression but not rituximab promotes anti-donor alloresponses associated with favorable outcomes. Clinical Trial Registration: Registered with EudraCT (2006-002330-38) and www.ClinicalTrials.gov, identifier: NCT00476164.

Keywords: B lymphocytes; chronic rejection in renal transplant; donor specific antibody (DSA); kidney transplantation; rituximab.

Figure 1

Consort diagram for RituxiCAN-C4 trial. *Indicates 47 patients included in the exploratory analysis. ^†According to pre-specified second interim per protocol analysis.

Figure 2

Response to optimized immunosuppression. Exploratory analysis comparing those who responded favorably to optimized IS with those who did not. Graphs are box and whisker plots showing median with interquartile range (IQR) with whiskers showing upper and lower limits and outliers indicated as single data points. Means are represented with “x.” Time points: 0, enrolment sample; EP-1, End phase 1; 0–36, months post enrolment. White bars (n = 15); patients who responded well to optimized IS. Gray bars (n = 31 pre-enrolment. n = 32 post (one recruit did not have sufficient pre-enrolment creatinines); patients who failed to respond to optimized IS. (A) Urine PCR, (B) Tacrolimus trough levels, (C,D) Systolic, (C) and diastolic, (D) blood pressure (BP), (E,F) ΔeGFR normalized to enrolment ΔeGFR of 0. (G) Changes in Median Fluorescence Intensity (MFI) of cumulative DSA with time (NB includes values where DSA = 0). P-values by Mann Whitney U-test.

Figure 3

Changes in B cells with rituximab—data from RCT per protocol groups. Graphs are box and whisker plots showing median with interquartile range (IQR) with whiskers showing upper and lower limits and outliers indicated as single data points. Means are represented with “x.” Time points: 0= enrolment sample. EP-1, End phase 1; EP-2, End phase 2; 0–36, months post enrolment. Rituximab administered between EP-1 and EP-2. The gating strategy is described in detail in methods. “N” refers to the number of samples at each time point. (A–F) Changes in B cells in RCT. (A) Absolute numbers of B cells per uL of serum. (B–F) Flow cytometric analysis of the proportion of B cell subpopulations against time. (B) CD27-negative B cells as proportion of total CD19+ cells. (C) CD27+ B cells as proportion of CD19+ cells. (D) CD38loCD24lo cells as proportion of CD27- cells (naïve B cells). (E) CD38++CD24++cells as proportion of CD27- cells (Transitional T1 cells). Median absolute number of T1 per μL is shown beneath each column. (F) CD38+CD324+ cells as proportion of CD27- cells (Transitional T2 cells). P-values by Mann Whitney U-test.

Figure 4

ELISPOT patterns. (A–E) Illustrates the 3 basic patterns of anti-donor IFNγ production, displayed as the spot count (corrected for flow cytometric assessment of CD4+ cell proportions) present under 4 different conditions: CD8- (CD8-depleted PBMC); CD19- (CD8- & CD19-depleted PBMC), both performed in presence or absence of CD25+ cells. Samples showing anti-donor responsiveness from 51 recruits, including from those not in the exploratory analysis, are represented. (A) Pattern 1: Unregulated B cell-dependent pattern. Showing spot counts that reduce (>20%) on depletion of CD19+ cells in presence of CD25+ cells (± in absence of CD25+ cells). N = 16 samples. (B,C) Pattern 2: B cell-dependent anti-donor patterns with evidence of regulation. (B) CD25+ regulated B-dependent responses: B cell-dependent anti-donor responses only detectable in absence of CD25+ cells. N = 14 samples. (C) CD19+ regulated B-dependent responses. B cell-dependent anti-donor responses in presence of CD25+ cells, but when CD25+ cells absent, depletion of CD19+ cells increases spot count (>20%), indicating evidence of regulation by B cells. N = 2 samples. (D,E) Pattern 3: Regulated anti-donor responses without evidence of B cell-dependency. (D) CD19+ regulated responses; In presence of CD25+ cells, spot counts increase (>20%) when CD19+ cells are depleted. N = 11 samples. (E) CD25+ and CD19+ regulated. In absence of CD25+ cells, depletion of CD19+ cells increases spot counts (>20%). In presence of CD25+ cells, anti-donor responses are undetectable. N = 7 samples.

Figure 5

Associations with ELISPOT patterns. Graphs show box plots of median with IQR with whiskers showing upper and lower limits and outliers indicated as single data points. Means are represented with “x.” (A) Association between proportion of CD4+CD25+CD39hi T cells (Tregs) and ELISPOT patterns characterized by spot count suppression when CD25+ cells present. (B,C) Association between proportion of CD19+ cells having the phenotype of transitional T1 cells (CD27-CD38++CD24++) (B) or transitional T2 cells (CD27-CD38+CD24+) (C) and ELISPOT patterns showing evidence of increasing spot counts after depletion of CD19+ cells. (D,E) ΔeGFR, normalized to enrolment eGFR of 0 (D) and urine PCR (E) in patients with at least two samples at end-phase 2 or beyond (n = 27). White bars are patients who had either donor non-responsiveness or ELISPOT patterns with evidence of regulated anti-donor responses in their post-optimization samples (n = 21). Gray bars are those with at least one post-end-phase 2 sample showing evidence of unregulated B cell dependent anti-donor responses (n = 6). Time points: 0, enrolment sample; EP-1, End phase 1; EP-2, End phase 2; 0–36, months post enrolment. P-values by Mann Whitney U-test.