Visual Abstract
Keywords: acute allograft rejection, immunohistochemistry, kidney biopsy, molecular genetics, transplant outcomes
Abstract
Key Points
Stable kidney transplant recipients with early stage 1 lymphoid aggregates displayed gene signatures indicative of a tolerogenic immune environment.
Participants with early stage 1 lymphoid aggregates demonstrated a trend toward better kidney function trajectories and lower risk of graft failure.
Early stage 1 lymphoid aggregates may represent beneficial immune surveillance mechanisms promoting long-term graft stability in low-risk transplant recipients.
Background
The presence of lymphoid tissues and their effect on long-term kidney graft outcome remains unclear. This study investigates the occurrence, molecular and phenotypic characteristics of lymphoid aggregates (LA) in protocol biopsies of clinically stable kidney transplant recipients, and their correlation with long-term clinical outcome.
Methods
We retrospectively analyzed 414 protocol biopsies from a multicenter cohort of 278 stable kidney transplant recipients taken at 6, 12, or 24 months post-transplant. On a subset of 16 biopsies at 6 months post-transplant, mRNA gene expression analysis was performed using Nanostring Banff Human Organ Transplant panel alongside phenotypic analysis with immunostaining for developmental staging of LA. Finally, the presence of LA was correlated to various clinical kidney allograft outcomes.
Results
Stage 1 LA were characterized by upregulation of genes linked to immune regulation (LAG3, CTLA4, FOXP3), T-cell signaling, and chemotaxis pathways, with enrichment of regulatory T cells, memory T cells, T follicular helper cells, and dendritic cells. Immunohistochemistry confirmed higher numbers of FOXP3+ regulatory T cells and dendritic cells in stage 1 LA, while high endothelial venules were absent. Over follow-up, participants with stage 1 LA showed slightly higher eGFR values and a trend toward better graft survival compared with those without LA or with stage 2 LA, although most differences did not reach statistical significance.
Conclusions
In stable kidney transplant recipients, stage 1 LA displayed molecular and phenotypic features consistent with a regulated immune environment. The molecular and immunohistochemical analysis of stage 1 LA suggests a more regulatory molecular phenotype and composition, which may contribute to better long-term graft survival observed in this cohort of functionally stable transplant recipients. Observed trends toward more favorable outcomes suggest that early LA may not be deleterious and could represent local immune regulation, although larger prospective studies are needed to confirm these findings.
Introduction
Inflammation in tissues is typically marked by scattered immune cells. Under chronic inflammatory conditions—including aging, cancer, autoimmune diseases, and organ transplantation—immune cells can organize into lymphoid aggregates (LA), ranging from small, disorganized clusters of T and B cells to more structured tertiary lymphoid tissues (TLTs).1–3 In transplanted organs, chronic inflammation and the formation of lymphoid structures contribute to ongoing tissue injury, fibrosis, and gradual loss of allograft function.4 Understanding the mechanisms driving chronic inflammation and LA formation in kidney grafts may improve strategies for preserving long-term graft survival.
Sato et al. defined TLTs as de novo-formed lymphocyte aggregates during chronic inflammation.5,6 These structures form a specialized microenvironment that supports interactions among T and B cells, dendritic cells, and fibroblasts, differentiating them from scattered or nonorganized inflammatory infiltrates. TLTs progress through developmental stages that reflect kidney injury, with advanced stages indicating more severe inflammation and tissue damage.5 Thus, TLT maturity could serve as a surrogate marker of intragraft injury.5
LA, often observed in renal allograft biopsies, have traditionally been interpreted as features of chronic inflammation or evolving rejection. In chronic rejection, TLTs are linked to local B-cell activation, breakdown of tolerance, and intragraft production of donor-specific antibodies (DSA).7 Beyond their antibody-producing capacity, B cells may also promote injury through antigen presentation and cytokine secretion sustaining T-cell–mediated immune responses.8,9 Emerging evidence suggests not all LA within graft tissue are harmful. In operational tolerance—defined by stable, long-term graft function without immunosuppressive therapy—patients showed increased levels of memory B cells with regulatory profiles compared with participants with chronic rejection.10 Rosales et al. identified regulatory T cell (Treg)–rich tertiary lymphoid organs (TOLs) in accepted mouse kidney allografts.11 Unlike inflammatory tertiary lymphoid organs, TOLs lacked germinal centers and compartmentalization, characterized by an abundance of FOXP3+ Tregs, suggesting some lymphoid structures may promote tolerance rather than inflammation.
Many prior studies focused on LA in the context of rejection or failed grafts. Lee et al. provided first evidence of their presence in protocol biopsies of stable participants and their association with kidney allograft function.12 In their study, stage 2 TLTs were independently associated with progressive decline in graft function. Like Lee et al., our study focused on protocol biopsies from stable kidney transplant recipients enrolled in three large randomized controlled trials investigating immunosuppression weaning strategies.13–16 The aim of this study was to characterize LA at a phenotypic and molecular level in protocol kidney allograft biopsies from stable, low-risk recipients free from overt rejection or chronic inflammation. This descriptive approach allows exploration of LA composition and maturation and provides possible insights into their clinical relevance and potentially helps differentiate between harmful inflammation and beneficial immune regulation.
Methods
Study Design and Patient Cohort
The post hoc analysis included 278 adult kidney transplant recipients from the Amsterdam University Medical Center, University Medical Center Groningen, and Leiden University Medical Center, originally enrolled in the CAlciNeurin-inhibitor Nephrotoxicity and Efficacy Study 9 (CANNES), Low Tidal MEChanical Ventilation Against NO Ventilation During Cardiopulmonary Bypass Heart Surgery (MECANO), or Amsterdam, LEiden, GROningen (ALLEGRO) open-label, multicenter, randomized controlled trials.13,14,16 Ethical approval was obtained for each trial by the local ethics committee. Detailed descriptions of different study designs and randomization procedures are published elsewhere.13–16 Participants were between 18 and 74 years receiving a first or second kidney transplant from deceased or living donors. Inclusion required at least one adequate protocol biopsy meeting Banff 97 criteria (≥7 glomeruli and one artery).17 Exclusion criteria included HLA-identical donor transplants, panel-reactive antibodies >50%, or ABO incompatibility. Protocol kidney allograft biopsies were conducted at two predefined time points post-transplant: 6 and 12 months (CANNES), 6 and 24 months (MECANO), or 12 and 24 months (ALLEGRO). Follow-up clinical data were collected over a minimum of 5 years. A total of 414 protocol biopsies were included (Figure 1).
Figure 1.
Patient inclusion and biopsy flow chart. Flow diagram showing selection of kidney transplant recipients from three randomized controlled trials (CANNES, MECANO, and ALLEGRO) and the number of protocol biopsies available for analysis. Biopsies were stratified according to the presence or absence of LA, with subsequent classification of LA into stage 1 or stage 2 when adequate for immunohistochemical staging. ALLEGRO, Amsterdam, LEiden, GROningen; CANNES, CAlciNeurin-inhibitor Nephrotoxicity and Efficacy Study; FDCs, follicular dendritic networks; GC, germinal centers; LA, lymphoid aggregates; MECANO, Low Tidal MEChanical Ventilation Against NO Ventilation During Cardiopulmonary Bypass Heart Surgery.
Kidney Allograft Biopsies and LA Definition
Biopsies were formalin-fixed, paraffin-embedded and stained (hematoxylin and eosin, periodic acid–Schiff, and Jones) for evaluation by a single, blinded expert pathologist using the Banff classification.18–20 Instead of the term TLTs, this study used the broader term LA because these biopsies were taken from clinically stable participants with no rejection signs. This descriptive approach aimed to reflect the possibility of early immune regulation rather than pathology. LA were defined as clusters of 30+ mononuclear cells located in the interstitium, perivascular, or periglomerular areas without intervening tubules.21 Participants were grouped into those with LA in at least one biopsy versus those with no LA in either biopsy. Additional grouping was based on biopsy timing: 6, 12, or 24 months post-transplant.
Phenotypic Analysis and LA Staging
LA composition was evaluated using a sequential immunostaining protocol developed by De Boer et al.22 In short, biopsies were single-stained sequentially, with a digitalization and elution step after each staining round on the same tissue section. A total of four series of sequential stainings were performed using the following antibodies: BCL6 (anti-mouse Ab M7211, PG-B6p clone, DAKO), Kappa (anti-rabbit Ab, A0191, DAKO), Lambda (anti-rabbit Ab, A0193, DAKO), PNAd (anti-rabbit Ab IgM, MECA-79, BioLegend), D2-40 (anti-mouse Ab, M3619 DAKO), CD3 (anti-rabbit Ab, RM 1907, Sp7 clone, Thermo), CD20 (anti-rabbit Ab, M0755, Dako), Ki-67 (anti-rabbit Ab, RM 9106, Thermo), CD23 (anti-rabbit Ab, RM9123, Thermo), FOXP3 (anti-mouse Ab, Ab20034, 236A/E7 clone, Abcam), CD68 (anti-mouse Ab, M0876, PG-M1 clone, DAKO), CD123 (anti-mouse Ab, MONX11130, BR4MS clone, Monosan), DC-SIGN (anti-mouse Ab, MAB161, 120507 clone, Bio-Techne), LAG3 (anti-rabbit Ab, CST15372, D2G40 clone, Cell Signaling), and CTLA4 (anti-mouse IgG2a Ab, BSB2882, BsB-88 clone, BioSB). After each staining round, the slide was digitized, followed by antibody removal through a washing step before applying a new primary antibody. Images were merged into composite overlays using the Philips Image Management System and analyzed with ImageJ 1.53k. LA were staged on the basis of the staging criteria by Sato et al.5 In short, stage 1 consisted of intermixed T- and B-cell clusters, without follicular dendritic networks (FDCs) nor germinal centers (GC; CD23-/Ki-67-). Stage 2 contained aggregates with early stages of compartmentalization with FDCs, but lacked GC (CD23+/Ki-67+ scattered cells). Stage 3 consisted of fully developed aggregates with compartmentalization, FDCs and GCs (CD23+/Ki-67+ dense clusters of cells). Stage 2 and 3 were considered advanced stage. Stainings were analyzed and quantified per mm2 within the most representative aggregate.
Nanostring mRNA Analysis
Sixteen biopsies (eight with LA, eight without LA) taken at 6 months post-transplant were analyzed using the nCounter Banff Human Organ Transplant panel (see Supplemental Table 1).23 All LA were stage 1 without histological signs of rejection. Five 20 µm sections from formalin-fixed paraffin-embedded (FFPE) blocks were obtained from each biopsy for mRNA isolation. After xylene deparaffinization, mRNA was isolated using the Nucleospin Total RNA FFPE XS Isolation Kit (MN, item number 740969.50). mRNA quality was assessed with the Agilent TapeStation after which 50 ng of mRNA was incubated for 20 hours overnight. Data analysis was conducted using RStudio (version RStudio 2023.12.1+402 Ocean Storm Release for Windows 11). CIBERSORT deconvolution analysis was used to characterize cell composition within LA.24 See Supplemental Data for a comprehensive R script and used packages.
Immunosuppressive Regimens
Participants were treated with different area under the curve-controlled maintenance immunosuppressive regimens according to their respective study protocol. In the CANNES trial, participants were randomly assigned to either a tacrolimus-based or cyclosporine A (CsA)-based regimen post-transplant. In the MECANO trial, all participants were treated with basiliximab induction, oral prednisolone (P), mycophenolate sodium, and CsA during the first 6 months post-transplant. After this period, when the protocol biopsy was free from subclinical rejection, participants were randomized to dual therapy with prednisolone and CsA, mycophenolate sodium, or everolimus. For the ALLEGRO trial, all participants initially received induction therapy with basiliximab and methylprednisolone. Post-transplant they were randomly assigned to either the early steroid withdrawal group, treated with tacrolimus and mycophenolic mofetil (MMF) without prednisolone, or the standard immunosuppression group, treated with tacrolimus, MMF, and prednisolone. Supplemental Table 2 summarizes each immunosuppressive strategy; further protocol details are available in trial publications.13–15 Rejection episodes (biopsy-proven [BPAR] or borderline with clinical relevance) were managed per local protocols, typically with high-dose methylprednisolone. Steroid-resistant cases were treated with T-cell–depleting agents such as anti-thymocyte globulin or alemtuzumab.25
Primary and Secondary Clinical Outcomes
The primary clinical outcomes were death-censored graft failure, defined as return to dialysis or retransplantation, and eGFR trajectories calculated using the CKD Epidemiology Collaboration equation.26 Secondary outcomes included generation of DSAs, BPAR, and histological signs of graft inflammation and fibrosis as per individual Banff lesion scores.18,27 In CANNES, preformed DSA were measured using ELISA. The presence of DSA, either preformed or de novo, were measured using the Luminex single bead antigen assay in MECANO and ALLEGRO. See original studies for detailed descriptions.13,16
Statistical Analysis
Statistical analyses were conducted using RStudio (version RStudio 2023.12.1+402 Ocean Storm Release for Windows 11). Categorical outcomes were compared using the chi-squared test. Continuous variables were analyzed using t tests, Mann–Whitney U, or Kruskal–Wallis tests depending on data distribution and outcome. The Dunn test with Bonferroni correction was applied for multiple pairwise comparisons. Kaplan–Meier curves and the log-rank test were used for survival analysis. Hazard ratios and 95% confidence intervals (CI) for death-censored graft failure were calculated. Risk factors for the development of LA were identified with univariate regression analysis. eGFR trajectory over time was modeled using linear mixed effects models. Measured eGFR values (CKD Epidemiology Collaboration) were used as the dependent variable in the LMM; the term predicted eGFR refers to the model-fitted values from these measurements to illustrate estimated group trajectories over time. Owing to small sample sizes of stage 1 and 2 LA in each trial, statistical comparisons focused on No LA versus LA groups. Full statistical methods and R scripts are available in the Supplemental Data.
Results
Patient Characteristics
Baseline demographics and donor-related and transplant-related characteristics are shown in Table 1. Recipient, donor, and transplant characteristics of participants from the three individual trials are summarized in Supplemental Table 3.
Table 1.
Baseline characteristics of included participants
| Parameter | No LA | LA |
|---|---|---|
| Participants per cohort, N (%) | ||
| CANNES | 36 (18) | 9 (11) |
| MECANO | 70 (35) | 27 (34) |
| ALLEGRO | 93 (47) | 43 (55) |
| Number of protocol biopsies per cohort, N (%) | ||
| CANNES | 59 (18) | 12 (13) |
| MECANO | 121 (38) | 29 (32) |
| ALLEGRO | 143 (44) | 50 (55) |
| Number of adequatea protocol biopsy per protocol biopsy time, N (%) | ||
| T=6 mo post-transplant | 95 (29) | 29 (43) |
| T=12 mo post-transplant | 120 (38) | 22 (32) |
| T=24 mo post-transplant | 108 (33) | 17 (25) |
| Recipient characteristics | ||
| Sex, male/female, N (%) | 138/61 | 57/22 |
| (69/31) | (72/28) | |
| Hypertension, N (%) | 64 (43) | 31 (50) |
| Diabetes mellitus, N (%) | 22 (12) | 8 (10) |
| Smoking, N (%) | 79 (58) | 21 (43) |
| BMI (kg/m2), mean±SD | 26±5 | 25±4 |
| Age (yr), median (IQR) | 54 (42–65) | 53 (41–62) |
| Type of dialysis, N (%) | ||
| Pre-emptive | 44 (22) | 22 (28) |
| Hemodialysis | 62 (31) | 20 (25) |
| Peritoneal dialysis | 72 (36) | 30 (38) |
| Dialysis, type unknown | 17 (9) | 7 (9) |
| Total days on dialysis, median (IQR) | 809 (324–1448) | 806 (140–1358) |
| Cause of ESKD, N (%) | 21 (11) | 14 (18) |
| Chronic kidney failure NOS | ||
| Focal segmental glomerulosclerosis | 2 (1) | 2 (3) |
| GN | 48 (24) | 15 (19) |
| Polycystic kidney disease | 39 (20) | 15 (19) |
| CAKUT | 1 (1) | 5 (6) |
| Kidney vascular disease (including hypertension) | 43 (22) | 7 (9) |
| Diabetes mellitus | 8 (4) | 6 (8) |
| Pyelonephritis/interstitial nephritis | 10 (5) | 4 (5) |
| Othera | 27 (14) | 11 (14) |
| Donor characteristics | ||
| Sex, male/female, N (%) | 92/107 | 43/36 |
| (46/54) | (54/46) | |
| Age (yr), median (IQR) | 53 (43–61) | 50 (41–61) |
| Living/deceased, N (%) | 82/111 | 36/ |
| (42/58) | 42 (46/54) | |
| DBD/DCD, N (%) | 45/41 | 16/18 |
| (52/48) | (47/53) | |
| Transplantation characteristics | ||
| DGF, N (%) | 46 (26) | 20 (27) |
| HLA mismatches, mean±SD | 3±2 | 3±1 |
| Cold ischemia time, min, median (IQR) | 11 (3–16) | 10 (3–17) |
| Warm ischemia time, min, median (IQR) | 30 (22–36) | 30 (21–40) |
| Rejection history before first protocol biopsy, n (% of patients)c | ||
| Any BPAR | 11 (6) | 6 (8) |
| Acute TCMR 1 | 7 (4) | 4 (5) |
| Acute TCMR 2 | 2 (1) | 1 (1) |
| Acute AMR | 1 (1) | 1 (1) |
| Mixed acute rejection | 1 (1) | 0 |
| Induction therapy, N (%) | ||
| No induction | 0 | 0 |
| Anti-IL2 MoAb | 197 (100) | 79 (100) |
| rATG | 0 | 0 |
| Baseline immunosuppressive treatment, N (%) | ||
| Corticosteroids | 117 (82) | 48 (80) |
| CNI | ||
| Tacrolimus | 112 (63) | 9 (12) |
| CsA | 59 (33) | 50 (66) |
| No CNI | 8 (4) | 17 (22) |
| mTOR inhibitor | ||
| EVL | 18 (10) | 17 (22) |
| No mTOR inhibitor | 158 (90) | 61 (78) |
| Proliferation inhibitor | ||
| MMF-cellcept | 139 (79) | 55 (70) |
| No proliferation inhibitor | 37 (21) | 24 (30) |
ALLEGRO, Amsterdam, LEiden, GROningen; AMR, antibody-mediated rejection; Anti-IL2 MoAb, Anti-IL2, monoclonal antibody; BMI, body mass index; BPAR, biopsy-proven acute rejection; CAKUT, congenital anomalies of the kidney and urinary tract; CANNES, CAlciNeurin-inhibitor Nephrotoxicity and Efficacy Study; CNI, calcineurin inhibitor; CsA, Cyclosporine A; DBD, donation after brain death; DCD, donation after cardiac death; DGF, delayed graft function; EVL, everolimus; HLA, human leukocyte antigen; IQR, interquartile range; LA, lymphoid aggregates; MECANO, Low Tidal MEChanical Ventilation Against NO Ventilation During Cardiopulmonary Bypass Heart Surgery; MMF, mycophenolic mofetil; mTOR, mechanistic target of rapamycin; NOS, not otherwise specified; rATG, rabbit anti-thymocyte globulin; TCMR, TCMR, T-cell–mediated rejection.
An adequate protocol biopsy is defined as a protocol biopsy which meets the criteria as defined in the Banff classification on the basis of the hematoxylin and eosin stains and immunohistochemical stainings with which stage of the lymphoid aggregates could be determined.
Other kidney diseases include nephropathy due to drugs, cholesterol embolism, amyloidosis, cortical or tubular necrosis, Fabry disease, systemic sclerosis, gout, and other unidentified kidney disorders
Prior rejection indicates any biopsy-proven acute rejection occurring before the first protocol biopsy. Patients were counted once; when multiple episodes occurred, the classification reflects the most severe biopsy-proven rejection phenotype before the first protocol biopsy.
Early Stage 1 LA Were Most Frequently Detected at 6 months Post-transplant
Out of 278 participants, 79 (28%) had LA in their protocol biopsies, accounting for 91 biopsies. Adequate immunostaining allowed staging of LA in 51 of these participants (65%): 39 with stage 1 LA and 12 with stage 2 LA. The distribution of LA at different biopsy times is shown in Figure 2. Stage 1 LA was most frequently observed at 6 months post-transplant, whereas stage 2 LA seemed more often at 12 and 24 months. Similar patterns were observed within the individual trials (Supplemental Figure 1), although numbers per subgroup were too small for statistical comparisons.
Figure 2.
Prevalence of LA across protocol biopsy time points. Bar plots show the proportion of biopsies without LA and with LA at 6-, 12-, and 24-month post-transplant. Early post-transplant biopsies show high prevalence of stage 1 LA. Among LA-positive biopsies, only stage 1 and stage 2 were observed and are displayed separately. For this analysis, only biopsies with adequate immunostaining allowing staging of LA were included.
Stage 1 LA Showed Upregulation of Immune Regulatory and T-Cell Signaling Pathways
In stage 1 LA, upregulation of several genes related to T-cell regulatory function (LAG3, CTLA4), B-cell chemotaxis (CXCL13), T-cell migration (CXCL9, CXCR4), and endothelial activation (ANKRD1) were observed (Figure 3A). Genes involved in the T-cell receptor complex (CD3D, CD3E) and the cytotoxic T-cell coreceptor CD8A were also upregulated, consistent with activation of the T-cell receptor signaling pathway in gene set enrichment analysis (Figure 3E). Principal component analysis showed partial clustering of stage 1 LA biopsies compared with those without LA, although with considerable heterogeneity (Figure 3B). Immune cell deconvolution suggested relative enrichment of subsets including Tregs, CD4+ memory T cells, T follicular helper cells (Tfh), and M1 macrophages (Figure 3C). Interestingly, naïve B-cell signatures seemed enriched, whereas plasma cell signatures were suppressed despite the presence of Tfh cells. Pathway analysis indicated activation of IFNγ signaling, apoptosis, the complement system, and several pathways related to adaptive immune responses, T-cell receptor signaling, and immune regulation, including PD1 signaling (Figure 3, D and E).
Figure 3.
Gene expression signatures associated with stage 1 LA. (A) Volcano plot shows significantly upregulated and downregulated genes in stage 1 LA biopsies (N=8) compared with normal renal tissue biopsies without LA (N=8, control group). (B) PCA demonstrating partial separation of LA and control biopsies. Larger symbols (triangle and circle) represent the centroid positions of each group for visual reference. (C–E) GSEA using (C) CIBERSORT, (D) HALLMARK, and (E) REACTOME gene sets, highlighting immune cell subsets and signaling pathways activated or suppressed in stage 1 LA. GSEA, gene set enrichment analysis; PCA, principal component analysis; TCR, T-cell receptor.
Stage 1 LA was Enriched in T Cells and FOXP3+ Tregs Compared with Stage 2 LA
Stage 1 LA showed higher CD3/CD20 ratios compared with stage 2 LA, suggesting a more T-cell–dominant composition in early aggregates (Figure 4A and Figure 5, A and B). Stage 1 LA also exhibited higher numbers of FOXP3+ Tregs (median FOXP3+ T-cells/mm2: 726 in stage 1 versus 316 in stage 2 at T=24 months; P=0.03; Figures 4B and 5C). Presence of FOXP3+ cells correlated with stage 1 LA in early biopsies (T=6 months: r = –0.51, P = 0.01), whereas this association was weaker and not statistically significant at later time points (T=12 months: r = –0.37, P = 0.15; T=24 months: r = –0.20, P = 0.42). Trends toward higher prevalence of macrophages (CD68+), plasmacytoid dendritic cells (CD123+), and myeloid dendritic cells (DC-SIGN+) were also observed in stage 1 compared with stage 2 LA (Figure 4, C–E, and Figure 5D). No significant differences were seen in the presence of high endothelial venules (not shown), or in CTLA4 and LAG3 expression, between stage 1 and stage 2 LA (Figure 4, F and G).
Figure 4.
Immunophenotypic features distinguishing early and advanced LA. Immunohistochemical analysis comparing stage 1 (blue) and stage 2 (red) LA in protocol biopsies at 6-, 12-, and 24-month post-transplant. Panels show (A) T/B lymphocyte ratio (CD3/CD20), (B) FOXP3+ Tregs, (C) CD68+ macrophages, (D) CD123+ plasmacytoid dendritic cells (pDC), (E) DC-SIGN+ myeloid dendritic cells, and protein expression of (F) LAG3 and (G) CTLA4. Tregs, regulatory T cells
Figure 5.
Multiplex immunofluorescence overlay of stage 1 and stage 2 LA. (A) Sequential multiplex overlay image of a stage 1 LA showing CD3+ T cells (white), CD20+ B cells (green), and Ki67+ proliferating cells (red). Stage 1 infiltrates are predominantly composed of CD3+ T cells with minimal B cells and proliferation. (B) Sequential multiplex overlay of a stage 2 LA demonstrating increased CD20+ B cells (green) and Ki67+ proliferating cells (red) compared with stage 1, with a mixed T-cell and B-cell population. (C) Multiplex overlay of a stage 1 infiltrate including CD3 (white), CD123+ plasmacytoid dendritic cells (green), DC-SIGN+ myeloid dendritic cells (yellow), FOXP3+ Tregs (magenta), and CD68+ macrophages (red), highlighting the diversity and regulatory cell enrichment in early aggregates. (D) Multiplex overlay of a stage 2 LA showing CD123+ plasmacytoid dendritic cells (green) with CD3+ T cells (white), less DC-SIGN+ dendritic cells (yellow), FOXP3+ cells (magenta), and CD68+ macrophages (red).
Stage 1 LA Showed a Nonsignificant Trend toward Improved Graft Survival and Better eGFR Trajectories Compared with Stage 2 LA and No LA
The presence of LA within the first 2 years post-transplant was associated with a nonsignificant trend toward better long-term graft survival over 10 years (89% versus 79% in those without LA; P = 0.21; Figure 6A). When stratified, stage 1 LA seemed to follow more favorable survival trajectories compared with stage 2 LA or no LA, although none of these comparisons reached statistical significance (P = 0.36–0.83; Figure 6B). Within individual trial cohorts, no significant survival differences were observed (Supplemental Figure 2).
Figure 6.
Kidney allograft outcomes by presence and stage of LA. (A) Kaplan–Meier survival curves comparing death-censored graft survival in recipients with versus without LA during 10 years of follow-up. (B) Kaplan–Meier survival curves comparing no LA, stage 1 LA, and stage 2 LA. (C) LMM analysis of estimated glomerular filtration rate (eGFR, ml/min per 1.73 m2) over 5 years comparing no LA with LA. (D) LMM analysis of eGFR trajectories comparing no LA, stage 1 LA, and stage 2 LA. Predicted eGFR represents model-fitted values from the linear mixed model on the basis of measured eGFR data (CKD-EPI) used for visualization of estimated group trajectories over time. LMM, linear mixed model; CKD-EPI, CKD-Epidemiology Collaboration.
Over a 5-year follow-up, participants with LA overall had slightly higher predicted eGFR compared with those without LA (β=6.20, 95% CI, 0.55 to 11.86; P = 0.04; Figure 6C). When stratified, stage 1 LA showed higher predicted eGFR than no LA (β=7.82, 95% CI, 1.29 to 14.35; P = 0.02), whereas stage 2 LA did not differ (β=2.07, 95% CI, –7.25 to 11.38; P = 0.66; Figure 6D). In the individual cohorts, a significant difference was observed only in CANNES (P = 0.02), while MECANO and ALLEGRO showed no significant differences (P = 0.30 and P = 0.94; Supplemental Figure 2).
The Presence of LA was Not Associated with Differences in Acute Rejection or the Development of de novo DSA
Among participants without LA, 8 of 179 (6%) developed de novo DSA, compared with 4 of 73 (6%) in those with LA (P = 0.74). No significant differences were observed between groups for class 1 (P = 0.45) or class 2 DSA (P = 1.00).
No Significant Associations Were Found between LA and Risk of Acute Rejection or Immunosuppressive Treatment
Presence of LA during the first 24 months post-transplant was not associated with an increased risk of acute rejection during the subsequent 5 years (14% versus 13% in participants with versus without LA; P = 0.85). BPAR within the first 24 months did not increase the risk of LA development (P = 0.45). No significant risk factors for the development of stage 1 or 2 LA were identified (Table 2 and Supplemental Table 4). Treatment with CsA and early steroid withdrawal were associated with a lower prevalence of stage 1 LA, whereas tacrolimus and everolimus seemed to increase the prevalence of both stage 1 and stage 2 LA, although these differences were not significant (Table 3). In addition, no participants receiving P/CsA developed LA from 6 to 24 months, while five of 35 (14%) in the prednisolone/everolimus group did (P = 0.07). In the ALLEGRO trial, early steroid withdrawal was associated with fewer LA compared with standard regimens or tacrolimus minimization (P = 0.12) because of fewer stage 1 LA, while stage 2 LA prevalence was similar (P = 0.21).
Table 2.
Univariate analysis of risk factors for lymphoid aggregates development compared with no lymphoid aggregates in the first 2 years post-transplant
| Parameter | OR (95% CI)a | P Value |
|---|---|---|
| Recipient age (per yr increase) | 0.99 (0.97 to 1.01) | 0.29 |
| Recipient sex (male) | 0.87 (0.49 to 1.56) | 0.65 |
| Donor age (per yr increase) | 0.99 (0.97 to 1.01) | 0.18 |
| Donor sex (male) | 0.72 (0.43 to 1.22) | 0.22 |
| Recipient BMI | 0.98 (0.92 to 1.04) | 0.42 |
| HLA mismatches (per one mismatch increase) | 0.99 (0.84 to 1.17) | 0.93 |
| Diabetes mellitus type 2 (before transplantation) | 0.87 (0.37 to 2.05) | 0.76 |
| Cold ischemia time | 0.99 (0.96 to 1.02) | 0.54 |
| Warm ischemia time | 1.00 (0.98 to 1.02) | 0.87 |
| Delayed graft function | 1.06 (0.57 to 1.98) | 0.86 |
| Acute rejection (within first 6 mo post-transplant) | 1.18 (0.43 to 3.21) | 0.75 |
| Borderline TCMR (within first 6 mo post-transplant) | 0.70 (0.22 to 2.20) | 0.54 |
| Hypertension | 1.31 (0.72 to 2.38) | 0.37 |
| Smoking | 0.54 (0.28 to 1.05) | 0.07 |
BMI, body mass index; CI, confidence interval; HLA, human leukocyte antigen; OR, odds ratio; TCMR, T-cell–mediated rejection.
Stages are not considered; to analyze the risk of developing lymphoid aggregates, the participants with aggregates at 6, 12, and/or 24 months were pooled. The odds ratio presents the risk of developing lymphoid aggregates during the first 2 years post-transplantation.
Table 3.
Immunosuppressive regimens in three randomized trials and their relation to lymphoid aggregates development
| Cohort | No LA | LA | P Value |
|---|---|---|---|
| CANNES | |||
| T=6 | |||
| Tacrolimus | 23 (85%) | 4 (15%) | 1.00 |
| CsA | 15 (88%) | 2 (12%) | |
| T=12 | |||
| Tacrolimus | 22 (82%) | 5 (18%) | 0.38 |
| CsA | 16 (94%) | 1 (6%) | |
| MECANO | |||
| T=24 | |||
| P/CsA | 26 (100%) | 0 | 0.07 |
| P/EVL | 30 (86%) | 5 (14%) | |
| ALLEGRO | |||
| T=12 | |||
| Group 1: Early steroid withdrawal | 39 (89%) | 5 (11%) | 0.12 |
| Group 2: Normal immunosuppression | 38 (76%) | 12 (24%) | |
| Group 3: Tacrolimus minimization | 29 (71%) | 12 (29%) | |
| T=24 | |||
| Group 1: Early steroid withdrawal | 38 (86%) | 6 (14%) | 0.39 |
| Group 2: Normal immunosuppression | 44 (88%) | 6 (12%) | |
| Group 3: Tacrolimus minimization | 32 (78%) | 9 (22%) | |
CsA, cyclosporine A; EVL, everolimus; LA, lymphoid aggregates; P, prednisolone.
LA Were Associated with Higher Inflammation Scores but Not with Increased Graft Fibrosis
Participants with LA within the first 24 months post-transplant had higher total inflammation scores (ti-scores) compared with those without LA (P = 0.04, P = 0.02, and P = 0.02 at 6, 12, and 24 months, respectively; Table 4). The amount of graft fibrosis (interstitial fibrosis and tubular atrophy score) was similar between groups at all biopsy times, and increased inflammation did not correlate with more fibrosis development. Borderline inflammation did not differ between LA and No LA at 6 months (P = 0.81), 12 months (P = 0.09), or 24 months (P = 0.70), and subclinical rejection was not clinically treated.
Table 4.
Semiquantitative Banff lesion scores in participants with versus no lymphoid aggregates at different times post-transplant
| Total Cohort | |||||||
|---|---|---|---|---|---|---|---|
| Banff scorea | No LA | LA | P Value | Banff Score | No LA | LA | P Value |
| Graft inflammation | Graft fibrosis | ||||||
| t-score | IF/TA (%) | ||||||
| T=6 | 0 (0–1) | 0 (0–0.25) | 0.49 | T=6 | 10 (0–10) | 10 (0–10) | 0.42 |
| T=12 | 0 (0–0) | 0 (0–0.25) | 0.04 | T=12 | 10 (0–20) | 10 (0–15) | 0.83 |
| T=24 | 0 (0–0) | 0 (0–0) | 0.98 | T=24 | 10 (0–30) | 10 (10–40) | 0.10 |
| i-score | cg-score | ||||||
| T=6 | 0 (0–1) | 1 (0–1) | 0.01 | T=6 | 0 (0–0) | 0 (0–0) | 0.10 |
| T=12 | 0 (0–0) | 0 (0–1) | 0.01 | T=12 | 0 (0–0) | 0 (0–0) | 0.74 |
| T=24 | 0 (0–0) | 0 (0–1) | 0.09 | T=24 | 0 (0–0) | 0 (0–0) | 0.38 |
| g-score | cv-score | ||||||
| T=6 | 0 (0–0) | 0 (0–0) | 0.07 | T=6 | 0 (0–1) | 0.50 (0–1) | 0.14 |
| T=12 | 0 (0–0) | 0 (0–0) | 0.79 | T=12 | 0 (0–1) | 0 (0–1) | 0.89 |
| T=24 | 0 (0–0) | 0 (0–0) | 0.29 | T=24 | 0 (0–1) | 1 (0–2) | 0.01 |
| ti-score (%) | ah-score | ||||||
| T=6 | 10 (0–20) | 10 (0–30) | 0.04 | T=6 | 0 (0–1) | 0 (0–0) | 0.17 |
| T=12 | 10 (0–10) | 10 (0–30) | 0.02 | T=12 | 0 (0–1) | 0 (0–1) | 0.60 |
| T=24 | 10 (0–20) | 25 (10 35) | 0.02 | T=24 | 0 (0–1) | 1 (0–2) | <0.001 |
| v-score | |||||||
| T=6 | 0 (0–0) | 0 (0–0) | 0.57 | ||||
| T=12 | 0 (0–0) | 0 (0–0) | 0.59 | ||||
| T=24 | 0 (0–0) | 0 (0–0) | 0.49 | ||||
| ptc-score | |||||||
| T=6 | 0 (0–0) | 0 (0–0) | 0.54 | ||||
| T=12 | 0 (0–0) | 0 (0–0) | 0.01 | ||||
| T=24 | 0 (0–0) | 0 (0–0) | 0.11 | ||||
ah, arteriolar hyalinosis; Banff scores: g, glomerulitis; i, interstitial inflammation; ptc, peritubular capillaritis; t, tubulitis; ti, total inflammation; v, intimal arteritis; cg, glomerular basement membrane (GBM) double contours; cv, vascular fibrous intimal thickening; IF/TA, interstitial fibrosis and tubular atrophy; LA, lymphoid aggregates; mm, mesangial matrix expansion
Ordinal Banff scores are presented as median (interquartile range).
Discussion
Our study indicates stage 1 LA in stable, low-risk kidney transplant recipients may reflect a regulated immune microenvironment rather than a pathogenic one. This interpretation is supported by the upregulation of immune regulatory genes (FOXP3, LAG3, CTLA4) and the presence of naïve B cells lacking germinal center differentiation, features consistent with previous reports of regulatory T-cell and B-cell phenotypes in tolerant grafts.11,28 Although TLTs have historically been associated with chronic rejection, recent studies suggest that early or immature LA do not necessarily predict poor outcomes.12 Our molecular and phenotypic analyses of stage 1 LA are consistent with this refined view and support the concept that not all LA are immunologically deleterious. The transcriptional and cellular features observed in stage 1 LA may represent organized, but controlled, immune surveillance within the graft.
A key factor contributing may lie in the gene expression profiles and composition of stage 1 LA. Nanostring analysis demonstrated upregulation of genes related to T-cell regulatory functions and subsets involved in immune regulation, including Tfh cells, CD4+ memory T-cells, NK cells, and B-cells, similar to genes described by Rosales et al. in early accepted grafts.11 Furthermore, stage 1 LA were enriched for FOXP3+ Tregs and displayed other phenotypic features consistent with TOLS, identified as in situ markers of regulatory tolerance.11 Previous studies have emphasized the role of Tregs in maintaining tolerance in the allograft.29–31 However, it is important to note that Tregs are also upregulated in response to inflammation as a general physiological mechanism to regulate cytotoxic responses. Presence of FOXP3+ T cells may therefore not unequivocally reflect graft tolerance, but their presence in stage 1 LA could help stabilize local immune responses, potentially preventing progression to more advanced stages of LA and subsequent kidney injury.
Transplantation creates a highly immunologic environment in the graft, involving complex interactions between donor and recipient immune cells.32 B cells play a key role in this environment, producing DSA that mediate antibody-mediated rejection and graft loss.33 A positive correlation between graft function and TLT area in murine kidney allografts has been shown and identified IL-10+ CD19+ B cells as regulatory B cells in tolerant individuals, with reduced numbers in chronic rejection.34,35 In our study, stage 1 LA were enriched with naïve B cells that did not differentiate into plasma cells, suggesting these B cells were probably involved in immune regulation rather than antigen-driven immunity.36 Increased gene expression related to Treg function further supported the idea that these B cells play a role in immune modulation. Supporting this, Khiew et al. showed that in transplant tolerance, donor-specific B cells become functionally silenced and can suppress new alloreactive B cells. This highlights their potential as therapeutic targets to prevent antibody-mediated rejection.37 These findings raise the possibility that some LA or TLT-like structures may support tolerance rather than inflammation and challenge the view that intragraft lymphoid organization indicates harmful immune activation.
The presence of LA within the first 2 years post-transplant was associated with a nonsignificant trend toward better graft survival, with Kaplan–Meier curves diverging only after 5 years. This delayed separation may reflect gradual immunological or clinical effects but could equally be explained by changes in immunosuppression, comorbidities, or other unmeasured confounders during long-term follow-up. For kidney function, participants with stage 1 LA showed slightly higher predicted eGFR values compared with those without LA or with stage 2 LA, but the absolute differences were small, inconsistent across cohorts, and did not indicate distinct functional trajectories.
Tacrolimus has been reported to influence Treg biology by suppressing effector T-cell activation and proliferation through calcineurin inhibition, thereby reducing IL-2 consumption and indirectly supporting Treg survival.38,39 Clinical studies have shown higher FOXP3 and IL-10 expression and increased donor-specific Treg frequencies in tacrolimus-treated recipients, suggesting a cytokine milieu that favors regulatory over proinflammatory T-cell responses.40 These effects appear dose-dependent and context-dependent, but they may explain the relative enrichment of Tregs observed under tacrolimus-based immunosuppression.
Our study has several limitations. First, the retrospective and observational design, combined with the descriptive nature and limited numbers of biopsies, limits statistical power and the ability to infer causality. In addition, mRNA gene expression analysis was performed using the Banff Human Organ Transplant panel (770 selected genes), which covers only a small fraction of the transcriptome and may bias pathway and cell-type inference. Our molecular profiling was also performed on bulk FFPE sections, yielding a composite signal that cannot be assigned specifically to LA versus surrounding tissue. Moreover, LA-positive subgroups were too small to support robust analyses by aggregate burden (number/size) or location, and biopsy size variability together with the focal nature of LA increases susceptibility to sampling error. Future studies in larger cohorts are needed to compare stage 1 with more advanced LA to clarify mechanisms of progression and links to allograft outcomes. A further limitation is the lack of a universally accepted histologic definition for LA; although our criteria were based on prior studies, threshold-based definitions remain partly arbitrary and may affect interstudy comparability and reproducibility.
In stable kidney transplant recipients, early-stage LA showed molecular and cellular features indicative of an organized yet regulated immune milieu. Although these findings were largely descriptive and based on limited sample size, the trends observed suggest that early LA may not be deleterious and could represent local immune regulation. Larger, prospective studies integrating spatial and longitudinal analyses are needed to clarify whether these structures contribute to long-term graft stability or represent a transient, adaptive immune state.
Supplementary Material
Acknowledgments
Dr. Florquin and Dr. Kers designed the study. Mrs. Claessen performed practical experiments. Ms. du Long performed data collection and processing for this study. Dr. D. Roelen participated in practical experiments and data collection for the original trials. Mr. S. Meziyerh and J. van den Born provided data from the original trials. Dr. J. Sanders was involved as principal investigator for the original MECANO and ALLEGRO trials. All biopsies were revised by dr. Florquin. Ms. du Long wrote the original draft of the manuscript. Statistical analysis was performed by Ms. du Long together with Dr. Kers. The final version of the manuscript was revised and approved by all authors.
Footnotes
S.F. and J.K. contributed equally to this work.
Disclosures
Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/KN9/B592.
Author Contributions
Conceptualization: Frederike J. Bemelman, Sandrine Florquin, Jesper Kers, Dave L. Roelen, Jan-Stephan F. Sanders, Aiko P.J. de Vries.
Data curation: Jesper Kers, Dave L. Roelen, Aiko P.J. de Vries, Romy du Long, Joost C. van den Born.
Formal analysis: Jesper Kers, Soufian Meziyerh, Dave L. Roelen, Jan-Stephan F. Sanders, Romy du Long.
Funding acquisition: Sandrine Florquin.
Investigation: Frederike J. Bemelman, Nike Claessen, Sandrine Florquin, Jesper Kers, Soufian Meziyerh, Dave L. Roelen, Jan-Stephan F. Sanders, Aiko P.J. de Vries, Joost C. van den Born.
Methodology: Frederike J. Bemelman, Nike Claessen, Sandrine Florquin, Soufian Meziyerh, Dave L. Roelen, Jan-Stephan F. Sanders, Aiko P.J. de Vries, Romy du Long, Joost C. van den Born.
Resources: Sandrine Florquin.
Supervision: Sandrine Florquin, Jesper Kers.
Visualization: Jesper Kers, Romy du Long.
Writing – original draft: Sandrine Florquin, Jesper Kers, Romy du Long.
Writing – review & editing: Frederike J. Bemelman, Nike Claessen, Sandrine Florquin, Jesper Kers, Soufian Meziyerh, Dave L. Roelen, Jan-Stephan F. Sanders, Aiko P.J. de Vries, Romy du Long, Joost C. van den Born.
Funding
S. Florquin: Nierstichting (DKF15OP09).
Declarative Statements
This study includes clinical experimentation and received Institutional Review Board or Ethics Committee approval. All patients provided written informed consent. This study includes clinical experimentation and complies with the Declaration of Helsinki. The clinical and research activities being reported are consistent with the Principles of the Declaration of Istanbul as outlined in the Declaration of Istanbul on Organ Trafficking and Transplant Tourism.
Data Availability Statements
Original data generated for the study will be made available upon reasonable request to the corresponding author. Data Type: Image Data; Observational Data; Health Care Data; Research Protocols; Software Executable Code; Statistical Analysis Plan. Reason for Restricted Access: Due to the large size of the dataset and software code it will be available but on request.
Supplemental Material
This article contains the following supplemental material online at http://links.lww.com/KN9/B593.
Supplemental Table 1. Biopsy characteristics of included biopsies for the nanostring analysis.
Supplemental Table 2. Main study characteristics for the three different original study protocols (CANNES, MECANO, and ALLEGRO randomized controlled trials).
Supplemental Table 3. Baseline characteristics of included participants for the three individual trials.
Supplemental Table 4. Univariate analysis of risk factors for stage 1 and 2 LA within 24 months post-transplant.
Supplemental Figure 1. Prevalence of LA per trial.
Supplemental Figure 2. Kidney allograft survival and function by LA stage across trials.
References
- 1.Sato Y, Yanagita M. Immunology of the ageing kidney. Nat Rev Nephrol. 2019;15(10):625–640. doi: 10.1038/s41581-019-0185-9 [DOI] [PubMed] [Google Scholar]
- 2.Pitzalis C, Jones GW, Bombardieri M, Jones SA. Ectopic lymphoid-like structures in infection, cancer and autoimmunity. Nat Rev Immunol. 2014;14(7):447–462. doi: 10.1038/nri3700 [DOI] [PubMed] [Google Scholar]
- 3.Sato Y, Silina K, van den Broek M, Hirahara K, Yanagita M. The roles of tertiary lymphoid structures in chronic diseases. Nat Rev Nephrol. 2023;19(8):525–537. doi: 10.1038/s41581-023-00706-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dahle DO Mjøen G Oqvist B, et al. Inflammation-associated graft loss in renal transplant recipients. Nephrol Dial Transplant. 2011;26(11):3756–3761. doi: 10.1093/ndt/gfr163 [DOI] [PubMed] [Google Scholar]
- 5.Sato Y Boor P Fukuma S, et al. Developmental stages of tertiary lymphoid tissue reflect local injury and inflammation in mouse and human kidneys. Kidney Int. 2020;98(2):448–463. doi: 10.1016/j.kint.2020.02.023 [DOI] [PubMed] [Google Scholar]
- 6.Sato Y, Tamura M, Yanagita M. Tertiary lymphoid tissues: a regional hub for kidney inflammation. Nephrol Dial Transplant. 2023;38(1):26–33. doi: 10.1093/ndt/gfab212 [DOI] [PubMed] [Google Scholar]
- 7.Thaunat O Graff-Dubois S Fabien N, et al. A stepwise breakdown of B-cell tolerance occurs within renal allografts during chronic rejection. Kidney Int. 2012;81(2):207–219. doi: 10.1038/ki.2011.317 [DOI] [PubMed] [Google Scholar]
- 8.Nasr IW Reel M Oberbarnscheidt MH, et al. Tertiary lymphoid tissues generate effector and memory T cells that lead to allograft rejection. Am J Transplant. 2007;7(5):1071–1079. doi: 10.1111/j.1600-6143.2007.01756.x [DOI] [PubMed] [Google Scholar]
- 9.Zarkhin V Kambham N Li L, et al. Characterization of intra-graft B cells during renal allograft rejection. Kidney Int. 2008;74(5):664–673. doi: 10.1038/ki.2008.249 [DOI] [PubMed] [Google Scholar]
- 10.Pallier A Hillion S Danger R, et al. Patients with drug-free long-term graft function display increased numbers of peripheral B cells with a memory and inhibitory phenotype. Kidney Int. 2010;78(5):503–513. doi: 10.1038/ki.2010.162 [DOI] [PubMed] [Google Scholar]
- 11.Rosales IA Yang C Farkash EA, et al. Novel intragraft regulatory lymphoid structures in kidney allograft tolerance. Am J Transplant. 2022;22(3):705–716. doi: 10.1111/ajt.16880 [DOI] [PubMed] [Google Scholar]
- 12.Lee YH Sato Y Saito M, et al. Advanced tertiary lymphoid tissues in protocol biopsies are associated with progressive graft dysfunction in kidney transplant recipients. J Am Soc Nephrol. 2022;33(1):186–200. doi: 10.1681/ASN.2021050715 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bemelman FJ de Fijter JW Kers J, et al. Early conversion to prednisolone/everolimus as an alternative weaning regimen associates with beneficial renal transplant histology and function: the randomized-controlled MECANO trial. Am J Transplant. 2017;17(4):1020–1030. doi: 10.1111/ajt.14048 [DOI] [PubMed] [Google Scholar]
- 14.Rowshani AT Scholten EM Bemelman F, et al. No difference in degree of interstitial Sirius red-stained area in serial biopsies from area under concentration-over-time curves-guided cyclosporine versus tacrolimus-treated renal transplant recipients at one year. J Am Soc Nephrol. 2006;17(1):305–312. doi: 10.1681/ASN.2005030249 [DOI] [PubMed] [Google Scholar]
- 15.van Sandwijk MS de Vries APJ Bakker SJL, et al. Early steroid withdrawal compared with standard immunosuppression in kidney transplantation - interim analysis of the amsterdam-leiden-groningen randomized controlled trial. Transplant Direct. 2018;4(6):e354. doi: 10.1097/txd.0000000000000794 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.van den Born JC Meziyerh S Vart P, et al. Comparison of 2 immunosuppression minimization strategies in kidney transplantation: the ALLEGRO trial. Transplantation. 2024;108(2):556–566. doi: 10.1097/tp.0000000000004776 [DOI] [PubMed] [Google Scholar]
- 17.Racusen LC Solez K Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int. 1999;55(2):713–723. doi: 10.1046/j.1523-1755.1999.00299.x [DOI] [PubMed] [Google Scholar]
- 18.Loupy A Haas M Roufosse C, et al. The Banff 2019 Kidney Meeting Report (I): updates on and clarification of criteria for T cell- and antibody-mediated rejection. Am J Transplant. 2020;20(9):2318–2331. doi: 10.1111/ajt.15898 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Naesens M Roufosse C Haas M, et al. The Banff 2022 Kidney Meeting Report: reappraisal of microvascular inflammation and the role of biopsy-based transcript diagnostics. Am J Transplant. 2024;24(3):338–349. doi: 10.1016/j.ajt.2023.10.016 [DOI] [PubMed] [Google Scholar]
- 20.Naesens M Roufosse C Cornell LD, et al. The Banff 2024 Kidney Meeting Report: Rejection as a spectrum of phenotypes and focus on differential diagnostic reasoning [published online ahead of print March 5, 2026]. Am J Transplant. doi: 10.1016/j.ajt.2026.01.018 [DOI] [PubMed]
- 21.Scheepstra C Bemelman FJ van der Loos C, et al. B cells in cluster or in a scattered pattern do not correlate with clinical outcome of renal allograft rejection. Transplantation. 2008;86(6):772–778. doi: 10.1097/TP.0b013e3181860a74 [DOI] [PubMed] [Google Scholar]
- 22.de Boer OJ Krebbers G Mackaaij C, et al. Comparison of two different immunohistochemical quadruple staining approaches to identify innate lymphoid cells in formalin-fixed paraffin-embedded human tissue. J Histochem Cytochem. 2020;68(2):127–138. doi: 10.1369/0022155419897257 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mengel M Loupy A Haas M, et al. Banff 2019 Meeting Report: molecular diagnostics in solid organ transplantation-consensus for the Banff Human Organ Transplant (B-HOT) gene panel and open source multicenter validation. Am J Transplant. 2020;20(9):2305–2317. doi: 10.1111/ajt.16059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Newman AM Liu CL Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453–457. doi: 10.1038/nmeth.3337 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Koshy P Furian L Nickerson P, et al. European Survey on clinical practice of detecting and treating T-Cell mediated kidney transplant rejection. Transpl Int. 2024;37:12283. doi: 10.3389/ti.2024.12283 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Levey AS Stevens LA Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–612. doi: 10.7326/0003-4819-150-9-200905050-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bouatou Y Nguyen TQ Roelofs J, et al. A multicenter application of the 2018 banff classification for BK polyomavirus-associated nephropathy in renal transplantation. Transplantation. 2019;103(12):2692–2700. doi: 10.1097/tp.0000000000002712 [DOI] [PubMed] [Google Scholar]
- 28.Guinn MT Szuter ES Yokose T, et al. Intragraft B cell differentiation during the development of tolerance to kidney allografts is associated with a regulatory B cell signature revealed by single cell transcriptomics. Am J Transplant. 2023;23(9):1319–1330. doi: 10.1016/j.ajt.2023.05.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Adeegbe D, Bayer AL, Levy RB, Malek TR. Cutting edge: allogeneic CD4+CD25+Foxp3+ T regulatory cells suppress autoimmunity while establishing transplantation tolerance. J Immunol. 2006;176(12):7149–7153. doi: 10.4049/jimmunol.176.12.7149 [DOI] [PubMed] [Google Scholar]
- 30.Velásquez-Lopera MM Eaton VL Lerret NM, et al. Induction of transplantation tolerance by allogeneic donor-derived CD4(+)CD25(+)Foxp3(+) regulatory T cells. Transpl Immunol. 2008;19(2):127–135. doi: 10.1016/j.trim.2008.02.003 [DOI] [PubMed] [Google Scholar]
- 31.Miyajima M Chase CM Alessandrini A, et al. Early acceptance of renal allografts in mice is dependent on foxp3(+) cells. Am J Pathol. 2011;178(4):1635–1645. doi: 10.1016/j.ajpath.2010.12.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hsiao HM, Li W, Gelman AE, Krupnick AS, Kreisel D. The role of lymphoid neogenesis in allografts. Am J Transplant. 2016;16(4):1079–1085. doi: 10.1111/ajt.13645 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Zhang R. Donor-Specific antibodies in kidney transplant recipients. Clin J Am Soc Nephrol. 2018;13(1):182–192. doi: 10.2215/CJN.00700117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Nowocin AK, Meader L, Brown K, Elgueta R, Wong W. Characterizing the B-cell and humoral response in tertiary lymphoid organs in kidney allografts. Exp Clin Transplant. 2019;17(3):330–338. doi: 10.6002/ect.2017.0261 [DOI] [PubMed] [Google Scholar]
- 35.Silva HM Takenaka MC Moraes-Vieira PM, et al. Preserving the B-cell compartment favors operational tolerance in human renal transplantation. Mol Med. 2012;18(1):733–743. doi: 10.2119/molmed.2011.00281 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Peng B, Ming Y, Yang C. Regulatory B cells: the cutting edge of immune tolerance in kidney transplantation. Cell Death Dis 2018;9(2):109. doi: 10.1038/s41419-017-0152-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Khiew SH Jain D Chen J, et al. Transplantation tolerance modifies donor-specific B cell fate to suppress de novo alloreactive B cells. J Clin Invest. 2020;130(7):3453–3466. doi: 10.1172/jci132814 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Shahi A Salehi S Afzali S, et al. Evaluation of thymic output and regulatory T cells in kidney transplant recipients with chronic antibody-mediated rejection. Biomed Res Int. 2021;2021:6627909. doi: 10.1155/2021/6627909 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Paek JH, Kim YN, Shin HS, Jung Y, Rim H. Expansion and characterization of regulatory T cell populations from Korean kidney transplant recipients. Medicine (Baltimore). 2023;102(11):e33058. doi: 10.1097/md.0000000000033058 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Tsaur I Gasser M Aviles B, et al. Donor antigen-specific regulatory T-cell function affects outcome in kidney transplant recipients. Kidney Int. 2011;79(9):1005–1012. doi: 10.1038/ki.2010.533 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Original data generated for the study will be made available upon reasonable request to the corresponding author. Data Type: Image Data; Observational Data; Health Care Data; Research Protocols; Software Executable Code; Statistical Analysis Plan. Reason for Restricted Access: Due to the large size of the dataset and software code it will be available but on request.