Journal Logo

Reviews

The Implications of B-lineage Cells in Kidney Allografts

Filippone, Edward J. MD1,2; Farber, John L. MD3

Author Information
doi: 10.1097/TP.0000000000003163
  • Free

Abstract

INTRODUCTION

The majority of cells that comprise a kidney allograft cell-mediated rejection (CMR) are T cells and monocytes/macrophages, with B cells, plasma cells, and eosinophils typically accounting for <5%. Lymphocytes and monocytes also predominate in the interstitial infiltrates of antibody-mediated rejection (AMR), although neutrophils may be found in peritubular capillaries. B-lineage cells (see Figure 1), including various subsets of B lymphocytes and plasma cells, however, may constitute a significantly greater percentage of the cellular infiltrates of kidney allograft biopsies, whether performed for dysfunction or protocol. Whereas plasma cells are usually distributed diffusely, B cells often aggregate in nodules that may mature into tertiary lymphoid organs (TLOs) defined by expression of high endothelial venules (HEVs) and lymphatic vessels (LVs) characteristic of canonical secondary lymphoid organs (SLOs), such as lymph nodes and the spleen. Specific therapies are increasingly available that target B cells (anti-CD20 monoclonal antibodies), plasma cells (proteasome inhibitors and anti-CD38 monoclonal antibodies), and TLOs (costimulatory blockers and cytokine blockers among others). Thus, the pathogenic role of these infiltrates should optimally be established before the therapy because there may instead be a tolerogenic function. The purpose of this review is to analyze the available data to better inform clinicians regarding the potential benefits and harms of treatment should an allograft biopsy contain such infiltrates.

FIGURE 1.
FIGURE 1.:
Maturation of B2 lineage cells and available therapeutic agents. Maturation of hematopoietic stem cells (HSCs) to pro-B cells, pre-B cells, then to immature B cells within the bone marrow (BM) is antigen independent, driven by immunoglobulin gene rearrangement of VDJ segments of the heavy chain and VJ rearrangements of light chains to form complete immunoglobulin (Ig)M B cell receptors. These immature B cells exit the BM and transition (transitional B cells expressing IgD in addition to IgM) within the spleen into either marginal zone (MZ) B cells or more commonly to mature but naïve follicular (FO) B cells. Either MZ or FO B cells can mature into short-lived plasma cells (PCs) but only FO B cells can circulate to form germinal center reactions together with T-follicular helper cells and follicular dendritic cells within secondary lymphoid organs such as lymph nodes and the spleen in an antigen specific manner. Such activated FO B cells can mature into class-switched memory B cells or plasmablasts that in turn can mature into plasma cells residing in BM, spleen, and intestine. Note expression of CD19 from Pro-B cells through memory B cells and plasmablasts but not on PCs. CD20 (targeted by rituximab and others) is expressed starting from pre-B cells through plasmablasts, but not PCs. B cell–activating factor (BAFF; targeted by belimumab), a potent B-cell activator, interacts with the BAFF receptor (BAFFR, expressed on all B lineage cells except PCs) as well as the receptors B cell maturation factor (BCMA) and transmembrane activator, calcium modulator and cyclophilin ligand interactor (TACI), which are expressed on PCs. Memory B-cell survival, however, does not depend on BAFF signaling. Atacicept binds both BAFF and the prosurvival factor APRIL (capable of binding to TACI and BCMA but not the BAFFR). CD38 (targeted by daratumumab) is expressed on most B lineage cells including regulatory B cells of the transitional (CD19+CD24hiCD38hiCD27−) and memory (B10 cells, CD19+CD24hiCD38+CD27+) compartments and PCs, except perhaps naïve FO cells. APRIL, a proliferation induced ligand.

B CELL–RICH INFILTRATES

By the time renal allografts are irreversibly rejected, B cells are present within the graft. Studies from the mid-1970s showed that irreversibly rejected, explanted kidney allografts contain B lymphocytes that comprised anywhere from 15% to 73% of isolated infiltrates, whereas T cells comprised from 8% to 72%.1,2 Over the ensuing decades, the major focus was on the role of T cells in CMR. In 2003, a seminal paper by Sarwal et al3 first implicated a role for B cells in dysfunctional pediatric kidney allografts. Using DNA microarrays to assess gene expression patterns in biopsies from normal and dysfunctional grafts, 12 cases of 26 that had acute rejection (AR) were detected with an obvious B cell signature. Using immunohistochemical staining for CD20 in 20 untreated AR cases, large aggregates of B cells were found in 9. In a further analysis of 31 archived biopsy samples with AR, 9 additional cases had such aggregates. Comparing the clinical course of patients considered CD20 positive (defined as >275 CD20+/high power field [hpf] in the highest field) with those considered negative (<100/hpf in all fields, giving a ratio >2.5×), significantly greater steroid resistance and graft loss occurred in the positive cases. In a follow-up study, B cell clusters were found in 17 of 32 pediatric biopsies and were again correlated with steroid resistance and reduced graft survival.4

Subsequently, numerous studies assessed the significance in kidney allograft biopsies of B cell infiltrates for steroid resistance and graft survival, whether organized in a nodular fashion or scattered diffusely throughout the interstitium (see Figure 2). Results have been inconsistent, perhaps because of varying definitions of positivity, small sample sizes, differing immunosuppressive regimens and patient populations, and variable follow-up. In no study was the presence of B cell infiltrates correlated with AMR or C4d positivity. A critically important factor may be transplant duration. Einecke et al6 assessed 177 clinically indicated renal allograft biopsies and found that B cell–associated transcripts (BATs) and immunoglobulin transcripts (IGTs) were increased in both CMR and AMR and correlated with current/future function, inflammation, and interstitial fibrosis/tubular atrophy. However, both BATs and IGTs were strongly correlated with time, and their association with function was lost after such correction.

FIGURE 2.
FIGURE 2.:
Patterns of B cell infiltration in kidney allograft tissue stained for CD20. Note the diffuse pattern (A) and the more common nodular clusters (B–F) that may be perivascular (C and D), nonperivascular (B and E), or both perivascular and nonperivascular (F). Adapted from Ferdman et al,5 Transplantation 2014;98:766, with permission from Wolters Kluwer Health, Inc.

Table 1 outlines both positive and negative studies associating B-cell infiltrates with steroid resistance and/or graft survival. Kerjaschki et al7 found nodular B-cell infiltrates in approximately 10% of allograft biopsies. These infiltrates were enriched with podoplanin-positive new LVs, although not with the HEVs that define TLOs (see below). Survival was reduced compared to cases lacking nodular infiltrates, but duration of follow-up was shorter in the latter group.7 In a follow-up study from the same group using sequential protocol biopsies, however, graft function was actually better in cases containing LVs in their biopsies.18 Hippen et al8 studied 27 biopsies with Banff IA/B AR in the first year and found nodular CD20-positive infiltrates (specific criteria not defined) in 6 with significantly increased steroid resistance and reduced graft survival. Tsai et al9 stained 45 pediatric biopsies for CD20 and found that ≥3 CD20-positive cells/hpf significantly correlated with AR and ≥2/hpf correlated with reduced graft survival. Muorah et al10 found significantly reduced graft survival and increased steroid requirement in 15 pediatric biopsies with ≥300 CD20-positive cells/hpf compared with 35 with <300. Hwang et al11 found CD20-positive infiltrates (>275/hpf) in 23 of 54 adult biopsies (43%) with AR. The positive cases had significantly more steroid resistance, incomplete recovery, and reduced graft survival.

TABLE 1. - Studies finding a positive and no significant association between B-cell infiltrates/nodules and steroid resistance and/or graft loss
Author Number Criteria for B cell positivity Percent of biopsies positive Population Association: SR/GL Comment
Studies with positive association
 Sarwal et al3 52 ARs >275/hpf+ vs <100/hpf a 35% Pediatric P = 0.01/P < 0.001 CD20+ positivity did not correlate with C4d+
 Kerjaschki et al7 35 biopsies with nodular infiltrates and podoplanin + LV NA 10% Adult NA/appeared worse but follow-up not equal Nodules had >50-fold increase of LV density
 Hippen et al8 27 ARs (IA/B) CD20+ clusters, minimum number not specified 22% Adult P = 0.015/P = 0.024 16 of 21 without clusters had scattered CD20+ cells
 Tsai et al9 45 biopsies/31 patients ≥2 CD20+/hpf 45% of patients Pediatric NA/P = 0.02 ≥3 CD20+/hpf correlated with AR P = 0.0001
 Zarkhin et al4 32 ARs >275/hpf 52% Pediatric P < 0.0001/P = 0.0006 CD20+ occurred later
 Muorah et al10 48 patients >300/hpf 29% Pediatric P = 0.0015/P = 0.043 CD20+ had significantly more steroid courses
 Hwang et al11 54 patients with AR >275/hpf 43% Adult P = 0.001/P = 0.003 CD20= did not correlate with Banff grade or C4d+
Studies with no significant association
 Eikmans et al12 10 progressors to GL vs 18 nonprogressors CD79a >275/hpf a NA Adult NA/NS % B cell+ no different in group progressing to GL vs nonprogressors
 Doria et al13 14 ARs CD20+ nodules 50% B cell nodules Adult NS/NA No difference in creatinine
 Kayler et al14 120 biopsies ≥15 cell cluster, CD20+ from 5% to 90% of cells 59% had any cluster, 15% had CD20+ cells >50% of cluster cells, 4% had >275/hpf Adult NS/NS Banff grade I CMR more frequent in cluster+ vs cluster– P = 0.0051
 Bagnasco et al15 74 biopsies >100 cell cluster of CD20+ cells vs <50/hpf 30% cluster >100 cells; 9% cluster >200 cells; 5% cluster >275 cells a Adult NA/NS No difference in graft function
 Scheepstra et al16 54 biopsies >275 CD20+/hpf vs <100/hpf a 28% B cell cluster+ Adult NS/NS 2/15 cluster+ were pediatric
Cluster: >30 cells without tubules Scattered B cell scores also did not correlate
 Carpio et al17 110 biopsies CD20+ cells/mm3 NA Adult NA/NS CD20+/mm3 higher in CMR
No difference in function
aAim is to insure a difference of >2.5×.
AR, acute rejection; CMR, cell-mediated rejection; GL, graft loss; hpf, high power field; LV, lymphatic vessel; NA, not available; NS, nonsignificant; SR, steroid resistance.

By contrast, others found no significant associations. Eikmans et al12 in a case control study found no difference in percent of biopsies with >275 B cells/hpf between a group of patients with subsequent allograft loss from interstitial fibrosis/tubular atrophy and a group with preserved function (38% versus 35%). Doria et al13 found no difference in graft function comparing 7 patients with B-cell nodules and 7 without. Kayler et al14 defined lymphoid clusters (LCs) as ≥15 cells, with B cells representing 5%–90%. Of 120 biopsies, 71 had clusters, with 11 of 71 having >50% of cluster cells as B cells and 3 of 71 having >275 B cells/hpf. There was no difference in time to AR, steroid resistance, renal function, or graft survival between LC-positive and negative cases, and those with >50% CD20-positive cells or >275/hpf CD20-positive clusters fared no worse. Similarly, Bagnasco et al15 found LCs containing >100 CD20-positive cells in 34 of 74 biopsies with AR, including 11 with >200 and 4 with >275. No difference was found in graft function or survival over 4 y of follow-up. Scheepstra et al16 reported no significant difference in response to therapy or long-term outcome based on the presence of either LCs containing >275 CD20-positive cells or scattered CD20-positive cells compared to biopsies with <100 CD20-positive cells. Studying 110 for-cause biopsies, Carpio et al17 found that CD20-positive infiltrates correlated significantly with CMR, retransplantation, and HLA mismatches but not future graft function or graft survival. In a randomized trial of rituximab induction versus placebo, subsequent ARs in rituximab patients had significantly lower B cell scores than rejections in placebo patients but no difference in steroid resistance, subsequent function, or graft survival.19

It is not clear if the type of antibody induction therapy at the time of transplantation affects the subsequent development or characteristics of B-cell infiltrates. Many studies did not specifically comment on the type or frequency of induction between B cell–positive and –negative cases.4,7,10,13,15 Others used similar induction (anti-CD25 monoclonal antibody) for all patients.3,8,9,16 Hwang et al11 used basiliximab induction in 43% of their total population with no significant difference (P = 0.317) between CD20-positive (8/23) and CD-negative (15/31) cases. Kayler et al14 did find a significant difference in development of CD20-positive LCs based on induction therapy, as 79% (11/14) of patients with no induction, 75% (41/55) with thymoglobulin induction, and only 37% (19/51) with Campath induction had such clusters (P = 0.0001). In contrast, Gallon et al20 found no difference in B-cell infiltrates in the rejecting biopsies of 12 patients induced with Campath compared to rejecting biopsies of 6 receiving no induction. Comparing rituximab induction to placebo in an randomized controlled trials (RCT), van den Hoogen et al19 found significantly lower B-cell scores and complete absence of intragraft B-cell clusters in the rituximab group in subsequent AR biopsies but no difference in graft function at 2 y.

The conflicting results on the prognostic significance of B-cell infiltrates outlined above may be at least partially explainable by the opposing roles B cells play in the alloimmune response. On the one hand, they enhance rejection by presenting antigen to T cells, secreting proinflammatory cytokines,21 and maturing into plasma cells producing potentially destructive alloantibodies and autoantibodies.22 On the other hand, B cells can have a major role in dampening the immune response by developing a regulatory phenotype.

Regulatory B cells (Breg) are not a specific B cell lineage driven by a defined transcription factor, such as forkhead box P3 (FoxP3) for regulatory T cells (Treg).23 Rather multiple B-cell subsets may adopt a regulatory phenotype when appropriately stimulated by chronic inflammation, including even plasmablasts24 and plasma cells.25 Well-characterized human Breg include transitional B cells (CD19+CD 24hiCD38hiCD27−) and B10 and B10pro cells (CD19+CD24hiCD27+). Multiple studies in experimental animals indicate that Breg are induced by various proinflammatoty cytokines (eg, interleukin [IL]-1β, IL-6 and IL-21), as well as by CD40 ligation, TLR engagement, and/or B-cell receptor activation.23 Bregs are characterized primarily by their ability to secrete IL-10 but also IL-35 and transforming growth factor-β. The result is enhanced proliferation and activity of Tregs with concurrent suppression of Th1, Th2, and Th17 effector cells. In addition to regulatory cytokines, Bregs may inhibit cytotoxic T-cell activity by direct cell-to-cell contact through expression of PDL1 and FasL. Since MHC II–deficient B cells are not regulatory, cognate cell–cell interactions may be required as well for enhancing Treg activity. A complete discussion of Breg development, phenotype, and function is beyond the scope of this review and can be found in several recent reviews.23,26,27

Evidence exists for a potential beneficial role for Breg in human kidney transplantation. Viklicky et al28 compared 21 patients with early acute AMR to 43 with early acute CMR and found an inverse correlation between CD20 mRNA expression and graft survival in both groups. In a prospective study of 73 de novo kidney transplant recipients, Shabir et al29 found transitional B-cell frequency protective against AR (hazard ratio, 0.6; P = 0.03). In cases of biopsy-proven AMR, Shiu et al30 demonstrated evidence of B-cell suppression of T-cell alloreactivity in an in vitro interferon-γ enzyme-linked immune absorbent spot assay. CD19 depletion enhanced donor responsiveness in approximately two-thirds of donor nonresponsive samples but also reduced responsiveness in donor-responsive samples, suggesting the presence of both regulatory and effector B-cell populations.30

True kidney transplant tolerance is generally defined as stable renal function for at least 1 y off all immunosuppression. Several groups have studied such tolerant patients and have found an enhanced B-cell signature compared to control groups including stable patients on immunosuppression, patients with declining function, and normal.31-33 It remains uncertain whether the enhanced B-cell signature is the cause of the tolerance or the result of it, as well as what role immunosuppressive medications played in the controls.

The therapeutic dilemma facing clinicians is whether the presence of a significant B-cell infiltration, nodular or diffuse, on a renal allograft biopsy should prompt depletional therapy with rituximab or other monoclonal B-cell antibodies (see Figure 1). As outlined above, the data are conflicting on the pathogenic significance of B-cell infiltrates (see Table 1). The majority of positive studies were conducted in pediatric patients, and the majority of studies in adults were negative. A meta-analysis of 5 of the studies involving 200 patients (117 pediatric) found a significant association with steroid resistance and graft loss, suggesting B-cell depletional therapy should be considered.34 Case reports also suggest a benefit to rituximab in treatment of refractory ARs involving diffuse35 or nodular36 B-cell infiltrates. Rituximab may deplete B cells within the allograft. Steinmetz et al retrospectively studied 16 patients with vascular rejections and found 9 had nodular B cell clusters (defined as >30 B cells in direct contact) of which 8 were treated with rituximab plus conventional therapy and 8 with conventional therapy alone. Subsequent biopsies in 14 showed complete depletion of B cells in the rituximab group and a slight increase in the conventional group. Both groups, however, had nearly identical responses to treatment.

In contrast, Thaunat et al37 demonstrated nodular B-cell persistence in 2 explanted failed kidney allografts following rituximab therapy despite profound peripheral B-cell depletion, attributed to enhanced expression of the B-cell survival factor B cell–activating factor (BAFF). This suggests the potential for combining anti-BAFF therapy (eg, belimumab) with rituximab in targeting potentially destructive nodular B-cell infiltrates and studies using anti-BAFF therapy are appearing. An ongoing study (NCT02500251) is assessing the use of belimumab in addition to rituximab, bortezomib, and plasmaphoresis as a pretransplantation desensitization strategy. In a murine model of kidney transplantation, an anti-BAFF monoclonal antibody combined with low-dose cyclosporine significantly reduced naïve B cells in allografts and the spleen.38 In a small phase 2 trial comparing belimumab to placebo in 28 patients receiving tacrolimus, mycophenolate, and steroids, no significant difference in naïve B cells was detected,39 although the IL-10:IL-6 ratio was increased (suggesting an enhanced regulatory role) and plasmablasts were reduced. It remains to be determined whether detection of a Breg signature should be sought (eg, by IL-10 staining) before consideration of anti-BAFF therapy.

A potential complication to rituximab, however, is that depletion of tolerizing B cells may enhance the alloimmune response and result in or worsen acute and chronic rejection. A randomized controlled trial in nonsensitized renal transplants comparing rituximab with daclizumab induction was suspended when 5 of 6 rituximab-treated patients developed biopsy-proven AR within 3 mo.40 In a nonrandomized trial of rituximab induction in HLA-incompatible kidney transplantation, a trend for higher AR in the first 3 mo occurred in the rituximab group (76% versus 56%) with more than double the number of CMRs (12/25 versus 5/25),41 although not all studies of rituximab induction have shown higher AR rates.42 Interestingly, rituximab may enhance development and activity of autoimmune diseases in both the experimental animal (autoimmune encephalomyelitis43) and in humans, including psoriasis44 and ulcerative colitis,45 presumably by depleting Breg.

Based on these concerns and the available conflicting observational data outlined above, we come to the following conclusions regarding the use of B-cell depleting therapy for diffuse or nodular B-cell infiltrates realizing that clinical equipoise clearly still exists (see Table 2). In well-functioning grafts with incidentally detected (protocol biopsy) B-cell infiltrates in the presence or absence of subclinical Banff-defined CMR or AMR, rituximab is not indicated, and treatment of the underlying subclinical rejection, if present, can be addressed. In acutely dysfunctional grafts satisfying Banff criteria for active AMR and/or CMR (acute or chronic active) but having B-cell infiltrates, therapy should be directed to the underlying type of rejection, as would be given in the absence of B-cell infiltrates. If there is resistance to conventional treatment with progressive graft decline, rituximab would be a consideration, but potential risks and benefits should be weighed on an individual basis. There are no available RCT on which to base a firm recommendation. If used, the first dose should probably be administered when the rejection is deemed resistant. The optimal regimen remains to be determined (eg, 4 weekly doses of 375 mg/m2 versus 1 g repeated in 14 d) as peripheral B-cell depletion is not necessarily mirrored within tissues. We would not recommend rituximab for acutely dysfunctional grafts with B-cell infiltrates not satisfying Banff criteria for CMR and AMR but having other pathology to explain the deterioration, for example, BK nephropathy, calcineurin inhibitor toxicity, or recurrent/de novo glomerulonephritis unless otherwise indicated (eg, membranous nephropathy).

TABLE 2. - Suggestions for treatment of B-cell infiltrates with anti-CD20 monoclonal antibodies
B-cell infiltrates found on protocol biopsy
 No specific anti-B cell therapy (anti-CD20 monoclonal antibody) indicated
 Treatment of any subclinical Banff-defined CMR and/or AMR as would be done in the absence of B-cell infiltrates
B-cell infiltrates found on for-cause biopsy
 Treatment of Banff-defined CMR and/or AMR as would be done in the absence of B-cell infiltrates
 If underlying rejection resistant to conventional therapy, consider anti-CD20 monoclonal antibody
 If other pathology found (BKVN, CNI toxicity, recurrent/de novo GN), treat accordingly without specific anti-B cell therapy
AMR, antibody-mediated rejection, active or chronic active; BKVN, polyoma BK virus nephropathy; CMR, T-cell-mediated rejection, acute or chronic, including borderline; CNI, calcineurin inhibitor; GN, glomerulonephritis.

TLOs ON ALLOGRAFT BIOPSY

Persistence of an antigenic challenge, as may occur in chronic infection, autoimmune disease, cancer, atherosclerosis, and transplantation, may result in generation and maturation of lymphoid nodules into TLOs, also known as ectopic lymphoid structures, in a process termed lymphoid neogenesis. These TLOs are characterized by a central core of B cells, with or without follicular dendritic cells (FDCs), T follicular helper cells (Tfhs), and germinal center (GC) formation, surrounded by a rim of T cells. Plasmablasts and plasma cells may be found around these clusters. Also present are newly formed, podoplanin-positive efferent LVs that serve as a source of egress for lymphocytes. In the kidneys, LVs normally accompany arteries and veins down to the interlobular artery level. They are not normally found in the interstitium. Lymphatic neoangiogenesis occurs in up to two-thirds of kidney allografts having nodular or diffuse cellular infiltrates.18 The defining feature of TLOs is the presence of HEVs, defined by a cuboidal endothelium and expression of peripheral nodal addressin, that represent a source of ingress of circulating mature but naïve lymphocytes. In contrast to SLOs like lymph nodes and the spleen, TLOs lack a capsule and a mantle zone and do not have an afferent lymphatic connection. Such TLOs may be found in kidney, lung, and heart allografts.

Both cell-mediated and humoral responses can be generated in TLOs. In the fields of transplantation and autoimmune diseases, TLOs are generally considered to be noxious, whereas in cancer, they are often beneficial by propagating antitumor immune responses. Data are emerging in the field of solid organ transplantation; however, the TLOs in both experimental models and in humans may in fact be tolerogenic, and this is discussed below. Although most commonly studied in failed, explanted kidney allografts that have chronically rejected, even biopsies of acutely rejecting kidney allografts may contain LCs including Tfh cells, FDCs, and activated, proliferating B cells indicative of GC formation.46

A detailed description of the genesis of TLOs is beyond the scope of this article. Recent reviews are available.47-49 Briefly, SLOs develop in utero through hematopoetically derived CD4-positive CD3-negative lymphoid tissue inducer cells (LTi) that activate stromal lymphoid tissue organizer cells via lymphotoxin-α1β2 to secrete chemokines such as CXCL13 attracting B-cells and CCL21 attracting T cells to their respective compartments. In comparison, in chronically inflamed tissues, in addition to LTi cells, other inflammatory cells such as monocytes/macrophages, dendritic cells, NK cells, and T cells may secrete a variety of cytokines in addition to lymphotoxin-α1β2, including tumor necrosis factor (TNF)-α, IL-17, and IL-22, that can also activate lymphoid tissue organizer cells to generate TLOs.

Data clearly support the pathogenic potential of TLOs in solid organ transplantation. Transplantation of TLO-containing skin allografts into mice lacking all SLOs (incapable of mounting an alloimmune response) were promptly rejected, as were allografts transplanted 60 d later, indicating a memory response.50 Using a rat allogeneic aortic transplantation model, Thaunat et al51 microdissected adventitia and found dense LC containing actively proliferating GC B cells with adventitial endothelial cells adopting a HEV phenotype. Cultured adventitial cells produced antibodies against donor-specific major histocompatibility complex-1 molecules, significantly more so than lymph node cultures. Similar GC-like structures were found in chronically rejected explanted human heart and kidney allografts. In another study of 20 explanted failed human kidney allografts, Thaunat et al52 found nodular accumulations of B cells in 19, resembling both primary lymph node follicles (GC absent) and secondary follicles with expression of activation-induced cytidine deaminase (GC present) and plasma cells in the periphery. Supernatants of cultured cortical fragments produced alloantibodies with distinct specificities as compared to those found in the serum that originated from SLOs.

By contrast, studies in experimental animals indicate a potential protective role of TLOs, suggesting that TLOs may simply enhance the underlying immune response, be it rejection or tolerance. Brown et al53 used a fully allogenic mouse model transplanting kidneys from DBA/2 mice into C57BL/6 mice. In this model having innate tolerance, 40% of grafts were indefinitely accepted without rejection and 40% had chronic rejection. Well-defined lymphoid clusters satisfying TLO criteria were found starting at 14 days and increased thereafter. Increasing TLO area was significantly correlated with better graft function. Using the same model, Nowocin et al54 also found survival with good function at 45 d and greater TLO area significantly correlating with better function. Using an innovative model, Pedersen et al55 knocked in vascular endothelial growth factor-C into C57Bl/6 mice and transplanted their kidneys into BALB/c mice. Control kidneys were rapidly rejected, and the mice died within 5 to 7 days. When vascular endothelial growth factor-C was induced, however, lymphangiogenesis containing LVs within Treg-rich organized lymphoid structures was stimulated and resulted in delayed rejection (P < 0.0007). Previous work in a similar model demonstrated the critical dependence of Foxp3-positive Treg in maintaining Treg-rich organized lymphoid structures and the tolerant state.56 In nonhuman primate allotransplantation, drug discontinuation may result in ectopic intragraft GCs, which may be found in both rejecting and tolerant grafts.57

Similar studies in human kidney transplantation would be nearly impossible to perform because the large amount of tissue required could only be obtainable from explanted whole grafts that had chronically rejected. In well-functioning grafts that may harbor tolerizing TLOs, explantation would not occur and sufficient tissue is not likely to be obtained from needle biopsies. Evidence for nonpathogenic or even tolerizing TLOs exists in human kidney and other solid-organ transplantation. In a study of 29 chronically rejected, explanted human kidney allografts, Xu et al58 found that 27% harbored TLOs yet lasted twice as long as those lacking such structures. Those grafts with TLOs, however, had less detectable FoxP3-positive cells, suggesting that TLOs may merely be an epiphenomenon of increasing graft duration and not directly involved in graft prolongation. In heart transplantation, the Quilty lesion, defined as nodular lymphocytic infiltrates detected on endomyocardial biopsies, may have all the features of TLOs, including peripheral nodal addressin–positive HEVs and LVs and is not considered rejection nor does it warrant treatment, although there may be an association with cellular and/or humoral rejections.59,60 In human lung transplantation, TLOs (termed inducible bronchus–associated lymphoid tissue) were found not to be associated with significant acute or chronic rejection61 and may in fact be inversely correlated with AR.62

An obvious issue is the therapeutic implications of finding a TLO in a kidney allograft biopsy, as lesions reminiscent of TLOs may occur with AR. Potential therapeutic measures to disrupt their structure and/or prevent their formation have been and are being studied in autoimmune diseases. For example, lymphotoxin, a member of the TNF superfamily and expressed by LTi cells, may be targeted by TNF inhibitors. In patients with rheumatoid arthritis, the presence of TLOs on synovial biopsy correlated with reduced response to anti-TNF therapy, and in a subgroup of TLO-positive patients, clinical response to TNF inhibition correlated with regression of TLOs on rebiopsy.63

A fully mature TLO would contain a GC composed of FDCs, Tfh cells, and B cells. Tfh cells are characterized by CXCR5 positivity, directing them to B cell–rich areas along with expression of the transcription factor B-cell lymphoma. They also express inducible costimulator (ICOS), CD40 ligand (CD40L), programmed cell death protein 1 (PDI), and IL21. Tfh-B cell crosstalk is critical to somatic hypermutation, affinity maturation, and isotype switching within GCs of both SLOs and TLOs.57,64 Targeting the critical costimulatory pathways involved in this interaction, specifically ICOS-ICOS ligand, CD40L-CD40, IL21-IL21R, and BAFF, is actively under study in autoimmune diseases,65,66 and results are being published.67,68 Furthermore, TLOs in allografts frequently lack FDC-positive GCs, yet Tfh-like cells (lacking CXCR5 and B-cell lymphoma 6 characteristic of GC Tfh cells) attracted to chronic inflammation may still activate B cells outside of formal GCs via ICOS, CD40L, and IL21 and may be potential targets for such therapy.69 As noted above, B-cell nodules may persist despite peripheral B-cell depletion following rituximab,37 suggesting the potential benefit of adding Tfh-B cell inhibition to B-cell depletion.

In kidney transplantation, belatacept was shown to be less associated with de novo donor-specific antibody production than cyclosporin.70 Although initially designed to inhibit T-cell activation through costimulatory blockade, there is now evidence that belatacept has direct effects on B-lineage cells. Belatacept reduced expression of CD86 in stimulated B cells and reduced plasmablast differentiation and plasma cell antibody production independent of T cell help.71 Belatacept was shown to inhibit GC function by inhibiting Tfh-B cell crosstalk.71,72 In contrast, however, de Graav et al73 found belatacept was no more effective than tacrolimus in disrupting Tfh-B cell crosstalk and was in fact less effective in inhibiting Tfh-associated plasmablast formation.

It remains uncertain whether therapy specifically aimed at disruption of TLOs is indicated in kidney transplantation. In our opinion, if Banff criteria for CMR and/or AMR are satisfied, conventional therapy for such rejections containing TLOs should be implemented. In the absence of Banff-defined rejection, enhanced immunosuppression targeting B cells or Tfh-B cell crosstalk would not be indicated if TLOs are present, given their tolerizing potential.

PLASMA CELL–RICH AR

The potential pathogenic role of plasma cells in mediating chronic rejection was first noted in 2 studies of failed explanted kidney allografts in the mid-1970s.2,74 Schlüter et al74 posited that plasma cells were the true “effector cells” of chronic rejection, and Busch et al75 found that plasma cells comprised about 20% of the total lymphoid infiltrate, being the most common in 3 of 16 explanted kidneys and the second most common in 9 of the remaining 13.2 Except for occasional papers commenting on intragraft antibody producing cells or plasma cells,75,76 the major focus on CMR involved T cells until a 1991 biopsy study by Nádasdy et al77 found that plasma cells comprised 20% of the cellular infiltrate in chronically rejecting grafts, much higher than in AR or cyclosporine nephrotoxicity.

The first description of a plasma cell infiltrate mediating AR occurred in 1993 by David-Neto et al78 in which they reported 22 cases (of 880 transplants) of acute allograft dysfunction, with biopsies showing plasma cells comprising 69% of infiltrating cells in what they termed “acute interstitial nephritis of plasma cells.” The outcome was poor, with 15 of 22 returning to dialysis and a possible association with viral infections, and medication allergy was suggested. In 1999, Charney et al79 presented 27 cases of “plasma cell-rich acute rejection”. All cases satisfied 1997 Banff criteria for AR (15 grade I, 10 grade II, and 2 grade III) and had a plasma cell–rich infiltrate defined as >300 plasma cells/20 hpfs. No cases had evidence for AMR, and only 2 of 15 tested had any HLA antibodies found months after their biopsies. There was no evidence of viral infection except for low-level Epstein-Barr virus (EBV) positivity (2.7% of cells). Similar to David-Neto et al, outlook was poor, with 14 of 27 failing.

Subsequently, at least 14 case-series (see Table 3) described the features of plasma cell–rich acute rejections (PCARs). Most series involved predominantly adult patients,11,79-81,83-86,88-90 but some were restricted to pediatric patients4,87,91 or did not report age.82 Varying definitions were used, but most commonly PCAR was considered when plasma cells comprised ≥10% of infiltrating cells. In general, PCARs were found in 5% or less of allograft biopsies,80,81,84-86,88,91 although there were exceptions. Gärtner et al used a cutoff of ≥20% plasma cells and found 14% of 357 biopsies and 47 explants positive.82 Hwang et al11 stained 54 adult biopsies for CD38 (captures plasmablasts and plasma cells, but as shown in Figure 1 also B cell precursors, immature B cells, transitional B cells, memory B cells, Breg, natural killer cells, myeloid-derived suppressor cells, and even some T regulatory cells92,93) and used a cutoff of ≥30%; 25 of 54 (46%) were positive. Interestingly, 23 of 54 (43%) had >275 CD20+ cells/hpf, and 15 of 54 (28%) were positive for both CD38 and CD20. Similarly, Zarkhin et al4 studied 32 pediatric biopsies and found CD38-positive cells in 19 of 32 (59%) without specifying a minimum number; CD20-positive clusters (>275/hpf) were found in 17 of 32 (52%), and 14 of 32 had both cell types present. Using a cutoff of 10%, Dufek et al87 found 14 of 162 (9%) pediatric biopsies contained plasma cell infiltrates.

TABLE 3. - Plasma cell-rich acute rejection series
Author Number Age Percent of biopsies Criteria Duration a Underlying Banff diagnosis Outcome Comments
Nádasdy et al77 42 B NA NA % PCs NA “CR” NA Associates PCs with CR
David-Neto et al78 22 B A 2.5% NA 25 ± 23 mo NA 15/22 lost grafts First study in AR
Charney et al79 27 B A NA >300 PCs/20 hpf 12/27 <6 mo 15 grade IA/B, 10 grade IIA/B, 2 grade III 14/27 failed No evidence for humoral component. No viruses detected
15/27 >6 mo
Meehan et al80 19 B A 2% Not specified 22 mo median 1 borderline, 16 grade IA/B, 2 grade IIA 13/16 lost graft, 8 in 1 y 3/19 had PTLD plus AR and lost graft. 10 had microscopic hemorrhage
Desvaux et al81 14 B A 1.8% >10% 187 d median 3 borderline, 11 grade IA/B 13/14 steroid resistant, 40% 1 y GS 3/5 C4d+, 8/12 with antidonor antibodies
Gartner et al82 53 B/E A 14% ≥20% NA Acute and chronic per Banff Reduced GS univariate Associated with VR which was only independent variable for GS
Zarkhin et al4 32 B P NA CD38+, % not specified 24.7 ± 31 mo no GL 10 borderline, 20 grade IA/B, 2 grade IIA. 7 AMR Steroid resistance Not all CD38+ cells are plasma cells
73 ± 50 mo GL
Hwang et al11 67 B A NA CD38+ >30% NA 51 grade I, 16 grade II Reduced GS Not all CD38+ cells are plasma cells
Rogers et al83 59 B A 12% ≥10% NA NA Reduced GS univariate Associated with PRA and HLA-MMs
Gupta et al84 8 B A 3% ≥10% 31.9 ± 17.2 mo 5 grade IA/B, 2 grade IIb, 1 grade III Reduced GS Plasma cell tubulitis noted
Carpio et al17 110 B A NA NA NA NA No correlation between % CD138+ cells and GS %CD138+ correlated with DSA, PRA, AMR
Abbas et al85 50 B A 3% ≥10% 3.1 ± 2.55 y All grade IA/B Reduced GS if DSA+ 65% DSA+, 34% C4d+
Uppin et al86 7 B A 3% >10% 17 mo median All AMR Graft loss in 2/7 Selected by PCs >10% and AMR
Dufek et al87 14 B P 9% >10% 18.6 mo median 3 borderline, 5 grade IA/B, 4 grade IIA/B, 1 chronic, 1 unclassifiable TIN 71% GL Reduced function in surviving grafts
Hasegawa et al88 50 B A 0.5% ≥10% 605 d median 4 borderline, 14 CMR, 9 AMR, 13 mixed CMR/AMR Multivariable HR for GL 8 PCAR can occur with all Banff types AR
Nishimura et al89 12 B A NA >10% 1340 d median 10 grade IA/B, 2 grade IIA Reduced GS Increased Th2 profile
Hamada et al90 6 B A NA >10% 302 ± 234 d 1 borderline, 4 grade IA/B, 1 grade II, 2 concurrent AMR 1 lost graft All had concurrent or prior viral infection
Alhamoud et al91 7 B P 5% >10% 25 ± 14.8 mo 7 grade IB Graft loss 43% Some response to bortezomib
aFrom time of transplantation until biopsy, mean unless specified median.
A, adult, all or predominantly; AMR, antibody-mediated rejection; B, biopsy; CMR, cell-mediated rejection; CR, chronic rejection; DSA, donor-specific antibodies; E, explants; GL, graft loss; GS, graft survival; HLA-MM, human leucocyte antigen mismatches; hpf, high power field; NA, not available; P, pediatric; PC, plasma cell; PCAR, plasma cell rich acute rejection; PRA, panel reactive antibody.

Overall, PCARs tend to occur later than nonplasma cell–rich ARs, which occur typically within the first 3–6 months posttransplantation. Nádasdy et al found plasma cells (PCs) as a feature of chronic rejection not AR, and the mean time in David-Neto et al’s series was 25 mo. Charney et al noted that 55% of PCARs occurred after 6 mo, compared with 32% of nonplasma cell–rich cases. Similarly, 77% of Meehan et al’s cases developed after 6 mo. Other series found median or mean onset after 187 d,81 251 d indigenous/869 d nonindigenous Australians,83 32 mo,84 3.1 y,85 17 mo,86 18.6 mo,87 489–1384 d (based on Banff category of rejection),88 1340 d,89 302 d,90 and 25 mo.91 These long durations raise the question of whether plasma cells are merely accumulating over time as an epiphenomenon and are not primarily involved, an issue raised with BATs and IGTs6 noted above.

The earlier reports suggested that PCARs were a variant of CMR, although many studies looked for evidence of a humoral component in PCARs. In a study of explanted failed kidney allografts, 58% had strong and diffuse CD138-positive staining and 43% had C4d-positive staining in ≥25% peritubular capillaries, with 32% having positive staining for both.94 Charney et al79 found no evidence for a humoral component in their 27 cases with all satisfying Banff 97 grades I, II, or III for CMR. The 16 cases of Meehan et al were all grade I or II CMR, but 10 did have interstitial hemorrhage.80 Desvaux et al81 found interstitial hemorrhages in 4 of 14 cases and antidonor antibodies (HLA or endothelial cell) in 9 of 12 tested. In a study of 50 living-related kidney transplants with grade I Banff CMR, Abbas et al85 found donor-specific HLA antibodies (DSA) in 64% and positive C4d staining in 36%. Uppin et al86 reported 7 patients with ≥10% PCAR complicating biopsies satisfying Banff 2007 criteria for AMR (all C4d-positive, 6 with DSA). Dufek et al87 found DSA in 43% of 14 pediatric cases. Of 40 cases with ≥10% plasma cells, Hasegawa et al88 found 4 borderline CMRs, 14 CMRs, 9 AMRs or suspicious for AMR, and 13 mixed CMR/AMRs per Banff 2015. Hamada et al90 found that 2 of 6 PCARs had concurrent AMR90 and 5 of 7 PCARs reported by Alhamoud et al91 had DSA. It is now clear that plasma cell infiltrates can be found in biopsies, with the whole gamut of underling Banff-defined ARs present from borderline, to CMR, to AMR, to mixed.

Overall, the presence of PCARs portend a bad prognosis. Most studies report 6 mo to 1 y graft survivals of 40% to 60%,78-81 although 1 y graft survivals of 75% to over 85%85,88,89 have been found. It remains uncertain whether the worse prognosis of PCARs compared to nonplasma cell–rich CMR is due to this type of rejection per se, a concurrent humoral component, or whether it is mainly a function of a significantly later occurrence.

Although most studies outlined above focus on the involvement of 1 cell type, B cell or plasma cell, some have considered both in the same biopsy. Correlations were somewhat different for each cell. Zarkhin et al4 found CD20-positive clusters correlated with steroid resistance and graft loss, but not with C4d positivity or DSA. CD38-positive infiltrates correlated with steroid resistance, C4d positivity, and DSA but not with graft loss. Hwang et al11 found that CD20-positive infiltrates correlated with steroid resistance, incomplete recovery, and reduced graft survival, whereas CD38-positive infiltrates correlated with incomplete recovery and graft loss but not steroid resistance. When considered together, graft survival was only significantly reduced when both CD20 and CD38 were positive, whereas there was no difference with only 1 positive versus neither. Carpio et al17 studied 110 biopsies with either no rejection (40), CMR (50), or AMR (20) and found that CD138-positive plasma cell–rich infiltrates correlated with AMR, panel reactive antibody, and DSA, whereas CD20-positive infiltrates correlated with CMR; no correlation existed between CD138 and CD 20-positive infiltrates, and neither predicted graft loss.

The original report of David-Neto suggested an association of PCAR with infection. Some studies found no such association, but others support it. In a study comparing 15 patients with AR in the first year who had documented cytomegalovirus or EBV infections also within the first year (3 concurrent and 8 with prior infection) to 15 noninfected patients with AR, Aiello et al95 found significantly more plasma cells (170 ± 110 per 20 hpf versus 19 ± 9; P < 0.01) and CD79a cells, which include B and plasma cells (451 ± 332 versus 60 ± 37; P < 0.01), along with significantly reduced graft survival in the infected group. Kemény et al96 studied 20 biopsy-proven polyoma virus–associated nephropathy cases and found that 10 (50%) had ≥15% plasma cells in their cellular infiltrates, including plasma cell tubulitis. Notably, IgM-secreting plasma cells were predominant in 50% and coequal with IgG in another 20% of cases, suggesting IgM positivity may be a clue to underlying polyoma virus as an earlier report in uninfected PCAR found predominant IgG.79

In addition to infection, a major concern in approaching PCARs is differentiation from posttransplant lymphoproliferative disorder (PTLD). Meehan et al80 reported 19 cases of PCAR, 3 of which had PTLD. One fatal case had polyclonal plasmacytic hyperplasia of the allograft and grade IA rejection with multiple lymph nodes involved at autopsy. The other 2 cases demonstrated monoclonality with graft-limited disease treated by explantation. It is imperative to differentiate PTLD from severe AR as initial modification of immunosuppression would be dramatically different, that is, reduction/discontinuation versus intensification, respectively (see Table 4).97 Clinical clues to PTLD include EBV D+/R− transplantation, previous severe AR treated with depletional therapy, EBV DNAemia, fever, cytopenias, lymphadenopathy, and/or extranodal masses. Pathologically, tubulitis and intimal arteritis may be found in PTLD and are not specific for rejection.98 Pathologic features indicating PTLD include nuclear atypia, expansile nodular infiltrates, and serpiginous necrosis.98,99 Infiltrates containing the whole gamut of lymphocyte differentiation, such as immunoblasts, noncleaved cells, and large and small cleaved cells in addition to plasma cells, support PTLD.97-99 Monomorphic PTLD clearly occurs, however, and rarely may comprise plasmablasts100 or plasma cells.101,102

TABLE 4. - Differentiation of PTLD from B-lineage infiltrates in AR
Clinical clues to underlying PTLD
 Prior AR, especially if treated with lymphocyte depleting antibodies
 EBV+ to EBV− transplantation
 EBV DNAemia by PCR
 Fever
 Cytopenias
 Adenopathy
 Extranodal masses
 Encephalopathy
Pathological clues to underlying PTLD
 Expansile nodular infiltrates
 Nuclear atypia
 Serpiginous necrosis (as opposed to hemorrhagic or wedge-shaped infarcts suggesting vascular AR)
 Entire B lineage present: small lymphocytes, large lymphocytes, plasmablasts, plasma cells
 Monoclonality of plasma cells confirms PTLD but is not required a
  Κ or λ light chain restriction by immunoflouresence or immunohistochemistry (perform in all cases)
  VDJ gene rearrangement analysis by PCR of DNA extracted from tissue (if uncertain)
  EBV positivity in 50% or more of cells (not required for diagnosis as PTLD can be EBV−, especially late)
  EBER by in situ hybridization
  Immunohistochemistry for latent membrane protein if EBER not possible or equivocal
Nondifferentiating features found in both PTLD and AR
 Tubulitis
 Intimal arteritis
aMonomorphic variants are typically monoclonal and indicate PTLD, whereas polymorphic variants include B-cell nodules that may be monoclonal without necessarily indicating PTLD.
AR, acute rejection; EBER, Epstein-Barr encoded RNA; EBV, Epstein-Barr virus, PCR, polymerase chain reaction; PTLD, posttransplant lymphoproliferative disease.

Although polyclonality characterizes early PTLD lesions,103 monoclonality occurs in both polymorphic and monomorphic lesions. Monoclonality of a plasma cell infiltrate confirms the diagnosis of PTLD and should be assessed in suspicious cases by in situ hybridization for κ or λ light chain restriction. Additionally, monoclonality can be established by immunoglobulin gene receptor rearrangement studies (VDJ) by polymerase chain reaction (PCR) of DNA extracted from formalin-fixed tissue.80 B cells typically proliferate clonally in response to the chronic antigenic stimulation and immunosuppression following transplantation.104 In fact, nodular B-cell clones in rejecting kidney allografts have been shown to be monoclonal by analyzing their immunoglobulin heavy chain variable regions.5,105 Hence, in contrast to plasma cell infiltrates, monoclonality of nodular B-cell infiltrates is not diagnostic of PTLD. In situ hybridization for EBV-encoded RNA should be performed in suspicious cases because positivity in greater than 50% of cells supports PTLD (compared to <1%–10% of cells in AR).98,99 Alternatively, EBV positivity can be confirmed by immunohistochemistry for EBV latent membrane protein.106

After ruling out infection and PTLD, the approach to therapy for PCAR should be directed at the underlying Banff type and grade of rejection: CMR, AMR, or mixed. In the case of resistance to maximal conventional therapy for such rejections, antiplasma cell therapy with proteasome inhibition or possibly daratumumab (anti-CD38) may be considered. Reports have been published suggesting a potential benefit of bortezomib.91,107,108 Kwun et al93 reported a patient with combined heart–kidney transplants with refractory kidney plasma cell-rich combined CMR and AMR given daratumumab. Eight of 9 DSAs were dramatically reduced, and repeat biopsy showed a significant decrease in plasma cell infiltrate. Daratumumab given to nonhuman primates, however, resulted in decreased numbers of Breg and Treg with an increase in activated T cells,93 a result reminiscent of patients with multiple myeloma given daratmumab in which Breg and CD38-expressing Treg were depleted along with increases in cytotoxic and memory T cells.92 Additionally, plasma cells and plasmablasts themselves may have a regulatory role, for example, by secreting IL-10,109,110 and it is possible that the plasma cells in PCAR may be exerting a suppressive role in what would otherwise have been an even more aggressive rejection. Obviously, more research on the specific role of plasma cell infiltrates is required, and RCTs of antiplasma cell agents are necessary before generalized recommendations for therapy can be made.

CONCLUSIONS

B cell infiltrates can be found in a significant minority of kidney allograft biopsies. Although they may occasionally be diffuse, these infiltrates are usually nodular with the potential for maturation into TLOs. The significance of these infiltrates remains uncertain. Some studies predominantly in pediatric patients indicated significant associations with steroid resistance and reduced graft survival. Other studies predominantly in adults did not find such associations. At this time, treatment of a biopsy containing a B-cell infiltrate should be directed at underlying rejections satisfying current Banff criteria for CMR and/or AMR. Given the tolerizing potential of B cells, specific depletional therapy is not automatically indicated. Should resistance to conventional therapy occur, rituximab would be a consideration, but data do not exist on which to base a firm recommendation. Plasma cell infiltrates are usually diffuse and may accompany CMR and/or AMR. It is imperative to rule out concurrent viral infections and PTLD in cases with plasma cell rich–infiltrates. Treatment otherwise should again be directed at the underlying rejection per Banff criteria, with antiplasma cell therapy a consideration in resistant and deteriorating grafts. Obviously more research is necessary to distinguish destructive from tolerizing infiltrates, whether B cell or plasma cell, and RCTs of therapy are necessary.

REFERENCES

1. Strom TB, Tilney NL, Carpenter CB, et al. Identity and cytotoxic capacity of cells infiltrating renal allografts. N Engl J Med. 1975; 292241257–1263
2. Busch GJ, Schamberg JF, Moretz RC, et al. T and B cell patterns in irreversibly rejected human renal allografts. Correlation of morphology with surface markers and cytotoxic capacity of the isolated lymphoid infiltrates. Lab Invest. 1976; 353272–280
3. Sarwal M, Chua MS, Kambham N, et al. Molecular heterogeneity in acute renal allograft rejection identified by DNA microarray profiling. N Engl J Med. 2003; 3492125–138
4. Zarkhin V, Kambham N, Li L, et al. Characterization of intra-graft B cells during renal allograft rejection. Kidney Int. 2008; 745664–673
5. Ferdman J, Porcheray F, Gao B, et al. Expansion and somatic hypermutation of B cell clones in rejected human kidney grafts. Transplantation. 2014; 987766–772
6. Einecke G, Reeve J, Mengel M, et al. Expression of B cell and immunoglobulin transcripts is a feature of inflammation in late allografts. Am J Transplant. 2008; 871434–1443
7. Kerjaschki D, Regele HM, Moosberger I, et al. Lymphatic neoangiogenesis in human kidney transplants is associated with immunologically active lymphocytic infiltrates. J Am Soc Nephrol. 2004; 153603–612
8. Hippen BE, DeMattos A, Cook WJ, et al. Association of CD20+ infiltrates with poorer clinical outcomes in acute cellular rejection of renal allografts. Am J Transplant. 2005; 592248–2252
9. Tsai EW, Rianthavorn P, Gjertson DW, et al. CD20+ lymphocytes in renal allografts are associated with poor graft survival in pediatric patients. Transplantation. 2006; 82121769–1773
10. Muorah MR, Brogan PA, Sebire NJ, et al. Dense B cell infiltrates in paediatric renal transplant biopsies are predictive of allograft loss. Pediatr Transplant. 2009; 132217–222
11. Hwang HS, Song JH, Hyoung BJ, et al. Clinical impacts of CD38+ B cells on acute cellular rejection with CD20+ B cells in renal allograft. Transplantation. 2010; 89121489–1495
12. Eikmans M, Roos-van Groningen MC, Sijpkens YW, et al. Expression of surfactant protein-C, S100A8, S100A9, and B cell markers in renal allografts: investigation of the prognostic value. J Am Soc Nephrol. 2005; 16123771–3786
13. Doria C, di Francesco F, Ramirez CB, et al. The presence of B-cell nodules does not necessarily portend a less favorable outcome to therapy in patients with acute cellular rejection of a renal allograft. Transplant Proc. 2006; 38103441–3444
14. Kayler LK, Lakkis FG, Morgan C, et al. Acute cellular rejection with CD20-positive lymphoid clusters in kidney transplant patients following lymphocyte depletion. Am J Transplant. 2007; 74949–954
15. Bagnasco SM, Tsai W, Rahman MH, et al. CD20-positive infiltrates in renal allograft biopsies with acute cellular rejection are not associated with worse graft survival. Am J Transplant. 2007; 781968–1973
16. 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; 866772–778
17. Carpio VN, Noronha Ide L, Martins HL, et al. Expression patterns of B cells in acute kidney transplant rejection. Exp Clin Transplant. 2014; 125405–414
18. Stuht S, Gwinner W, Franz I, et al. Lymphatic neoangiogenesis in human renal allografts: results from sequential protocol biopsies. Am J Transplant. 2007; 72377–384
19. van den Hoogen MW, Steenbergen EJ, Baas MC, et al. Absence of intragraft B cells in rejection biopsies after rituximab induction therapy: consequences for clinical outcome. Transplant Direct. 2017; 34e143
20. Gallon L, Gagliardini E, Benigni A, et al. Immunophenotypic analysis of cellular infiltrate of renal allograft biopsies in patients with acute rejection after induction with alemtuzumab (campath-1H). Clin J Am Soc Nephrol. 2006; 13539–545
21. Shen P, Fillatreau S. Antibody-independent functions of B cells: a focus on cytokines. Nat Rev Immunol. 2015; 157441–451
22. 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; 812207–219
23. Rosser EC, Mauri C. Regulatory B cells: origin, phenotype, and function. Immunity. 2015; 424607–612
24. Matsumoto M, Baba A, Yokota T, et al. Interleukin-10-producing plasmablasts exert regulatory function in autoimmune inflammation. Immunity. 2014; 4161040–1051
25. Shen P, Roch T, Lampropoulou V, et al. IL-35-producing B cells are critical regulators of immunity during autoimmune and infectious diseases. Nature. 2014; 5077492366–370
26. Wortel CM, Heidt S. Regulatory B cells: phenotype, function and role in transplantation. Transpl Immunol. 2017; 41:1–9
27. Alhabbab RY, Nova-Lamperti E, Aravena O, et al. Regulatory B cells: development, phenotypes, functions, and role in transplantation. Immunol Rev. 2019; 2921164–179
28. Viklicky O, Hribova P, Volk HD, et al. Molecular phenotypes of acute rejection predict kidney graft prognosis. J Am Soc Nephrol. 2010; 211173–180
29. Shabir S, Girdlestone J, Briggs D, et al. Transitional B lymphocytes are associated with protection from kidney allograft rejection: a prospective study. Am J Transplant. 2015; 1551384–1391
30. Shiu KY, McLaughlin L, Rebollo-Mesa I, et al. B-lymphocytes support and regulate indirect T-cell alloreactivity in individual patients with chronic antibody-mediated rejection. Kidney Int. 2015; 883560–568
31. 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; 785503–513
32. Sagoo P, Perucha E, Sawitzki B, et al. Development of a cross-platform biomarker signature to detect renal transplant tolerance in humans. J Clin Invest. 2010; 12061848–1861
33. Newell KA, Asare A, Kirk AD, et al.; Immune Tolerance Network ST507 Study Group. Identification of a B cell signature associated with renal transplant tolerance in humans. J Clin Invest. 2010; 12061836–1847
34. Lu Y, Li B, Shen Q, et al. Effects of CD20+ B-cell infiltration into allografts on kidney transplantation outcomes: a systematic review and meta-analysis. Oncotarget. 2017; 82337935–37941
35. Alausa M, Almagro U, Siddiqi N, et al. Refractory acute kidney transplant rejection with CD20 graft infiltrates and successful therapy with rituximab. Clin Transplant. 2005; 191137–140
36. Lehnhardt A, Mengel M, Pape L, et al. Nodular B-cell aggregates associated with treatment refractory renal transplant rejection resolved by rituximab. Am J Transplant. 2006; 64847–851
37. Thaunat O, Patey N, Gautreau C, et al. B cell survival in intragraft tertiary lymphoid organs after rituximab therapy. Transplantation. 2008; 85111648–1653
38. Steines L, Poth H, Schuster A, et al. Anti-BAFF treatment interferes with humoral responses in a model of renal transplantation in rats. Transplantation. 2020; 1041e16–e22
39. Banham GD, Flint SM, Torpey N, et al. Belimumab in kidney transplantation: an experimental medicine, randomised, placebo-controlled phase 2 trial. Lancet. 2018; 391101402619–2630
40. Clatworthy MR, Watson CJ, Plotnek G, et al. B-cell-depleting induction therapy and acute cellular rejection. N Engl J Med. 2009; 360252683–2685
41. Jackson AM, Kraus ES, Orandi BJ, et al. A closer look at rituximab induction on HLA antibody rebound following HLA-incompatible kidney transplantation. Kidney Int. 2015; 872409–416
42. van den Hoogen MW, Kamburova EG, Baas MC, et al. Rituximab as induction therapy after renal transplantation: a randomized, double-blind, placebo-controlled study of efficacy and safety. Am J Transplant. 2015; 152407–416
43. Matsushita T, Yanaba K, Bouaziz JD, et al. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease progression. J Clin Invest. 2008; 118103420–3430
44. Dass S, Vital EM, Emery P. Development of psoriasis after B cell depletion with rituximab. Arthritis Rheum. 2007; 5682715–2718
45. Goetz M, Atreya R, Ghalibafian M, et al. Exacerbation of ulcerative colitis after rituximab salvage therapy. Inflamm Bowel Dis. 2007; 13111365–1368
46. de Leur K, Clahsen-van Groningen MC, van den Bosch TPP, et al. Characterization of ectopic lymphoid structures in different types of acute renal allograft rejection. Clin Exp Immunol. 2018; 1922224–232
47. Hsiao HM, Li W, Gelman AE, et al. The role of lymphoid neogenesis in allografts. Am J Transplant. 2016; 1641079–1085
48. Koenig A, Thaunat O. Lymphoid neogenesis and tertiary lymphoid organs in transplanted organs. Front Immunol. 2016; 7:646
49. Alsughayyir J, Pettigrew GJ, Motallebzadeh R. Spoiling for a fight: B lymphocytes as initiator and effector populations within tertiary lymphoid organs in autoimmunity and transplantation. Front Immunol. 2017; 8:1639
50. 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; 751071–1079
51. Thaunat O, Field AC, Dai J, et al. Lymphoid neogenesis in chronic rejection: evidence for a local humoral alloimmune response. Proc Natl Acad Sci U S A. 2005; 1024114723–14728
52. Thaunat O, Patey N, Caligiuri G, et al. Chronic rejection triggers the development of an aggressive intragraft immune response through recapitulation of lymphoid organogenesis. J Immunol. 2010; 1851717–728
53. Brown K, Sacks SH, Wong W. Tertiary lymphoid organs in renal allografts can be associated with donor-specific tolerance rather than rejection. Eur J Immunol. 2011; 41189–96
54. Nowocin AK, Meader L, Brown K, et al. Characterizing the B-cell and humoral response in tertiary lymphoid organs in kidney allografts. Exp Clin Transplant. 2019; 173330–338
55. Pedersen MS, Müller M, Rülicke T, et al. Lymphangiogenesis in a mouse model of renal transplant rejection extends life span of the recipients. Kidney Int. 2020; 97189–94
56. Miyajima M, Chase CM, Alessandrini A, et al. Early acceptance of renal allografts in mice is dependent on foxp3(+) cells. Am J Pathol. 2011; 17841635–1645
57. Kwun J, Manook M, Page E, et al. Crosstalk between T and B cells in the germinal center after transplantation. Transplantation. 2017; 1014704–712
58. Xu X, Han Y, Wang Q, et al. Characterisation of tertiary lymphoid organs in explanted rejected donor kidneys. Immunol Invest. 2016; 45138–51
59. Di Carlo E, D’Antuono T, Contento S, et al. Quilty effect has the features of lymphoid neogenesis and shares CXCL13-CXCR5 pathway with recurrent acute cardiac rejections. Am J Transplant. 2007; 71201–210
60. Szymanska S, Grajkowska W, Sobieszczanska-Malek M, et al. Prevalence of the Quilty effect in endomyocardial biopsy of patients after heart transplantation - from cellular rejection to antibody-mediated rejection? Pol J Pathol. 2016; 673216–220
61. Hasegawa T, Iacono A, Yousem SA. The significance of bronchus-associated lymphoid tissue in human lung transplantation: is there an association with acute and chronic rejection? Transplantation. 1999; 673381–385
62. Hruban RH, Beschorner WE, Baumgartner WA, et al. Depletion of bronchus-associated lymphoid tissue associated with lung allograft rejection. Am J Pathol. 1988; 13216–11
63. Cañete JD, Celis R, Moll C, et al. Clinical significance of synovial lymphoid neogenesis and its reversal after anti-tumour necrosis factor alpha therapy in rheumatoid arthritis. Ann Rheum Dis. 2009; 685751–756
64. Walters GD, Vinuesa CG. T follicular helper cells in transplantation. Transplantation. 2016; 10081650–1655
65. Corsiero E, Delvecchio FR, Bombardieri M, et al. B cells in the formation of tertiary lymphoid organs in autoimmunity, transplantation and tumorigenesis. Curr Opin Immunol. 2019; 57:46–52
66. Bombardieri M, Lewis M, Pitzalis C. Ectopic lymphoid neogenesis in rheumatic autoimmune diseases. Nat Rev Rheumatol. 2017; 133141–154
67. Cheng LE, Amoura Z, Cheah B, et al. Brief report: a randomized, double-blind, parallel-group, placebo-controlled, multiple-dose study to evaluate AMG 557 in patients with systemic lupus erythematosus and active lupus arthritis. Arthritis Rheumatol. 2018; 7071071–1076
68. Genovese MC, Combe B, Kremer JM, et al. Safety and efficacy of ixekizumab in patients with PsA and previous inadequate response to TNF inhibitors: week 52 results from SPIRIT-P2. Rheumatology (Oxford). 2018; 57112001–2011
69. Hutloff A. T follicular helper-like cells in inflamed non-lymphoid tissues. Front Immunol. 2018; 9:1707
70. Vincenti F, Rostaing L, Grinyo J, et al. Belatacept and long-term outcomes in kidney transplantation. N Engl J Med. 2016; 3744333–343
71. Leibler C, Thiolat A, Hénique C, et al. Control of humoral response in renal transplantation by belatacept depends on a direct effect on B cells and impaired T follicular helper-B cell crosstalk. J Am Soc Nephrol. 2018; 2931049–1062
72. Ezekian B, Schroder PM, Mulvihill MS, et al. Pretransplant desensitization with costimulation blockade and proteasome inhibitor reduces DSA and delays antibody-mediated rejection in highly sensitized nonhuman primate kidney transplant recipients. J Am Soc Nephrol. 2019; 30122399–2411
73. de Graav GN, Hesselink DA, Dieterich M, et al. Belatacept does not inhibit follicular T cell-dependent B-cell differentiation in kidney transplantation. Front Immunol. 2017; 8:641
74. Schlüter E, Lennert K, Bohle A. The immunological significance of cellular infiltrates in chronic rejection of human kidney transplants. Virchows Arch A Pathol Anat Histol. 1975; 3683191–204
75. Busch GJ, Schamberg JF, Moretz RC, et al. Four patterns of human renal allograft rejection: a cytologic and in vitro analysis of the infiltrate in 24 irreversibly rejected kidneys. Transplant Proc. 1977; 9137–42
76. Garovoy MR, Reddish MA, Busch GJ, et al. Immunoglobulin-secreting cells recovered from rejected human renal allografts. Transplantation. 1982; 331109–111
77. Nádasdy T, Krenács T, Kalmár KN, et al. Importance of plasma cells in the infiltrate of renal allografts. An immunohistochemical study. Pathol Res Pract. 1991; 1872–3178–183
78. David-Neto E, Ribeiro DS, Ianhez LE, et al. Acute interstitial nephritis of plasma cells: a new cause for renal allograft loss. Transplant Proc. 1993; 251 Pt 2897–899
79. Charney DA, Nadasdy T, Lo AW, et al. Plasma cell-rich acute renal allograft rejection. Transplantation. 1999; 686791–797
80. Meehan SM, Domer P, Josephson M, et al. The clinical and pathologic implications of plasmacytic infiltrates in percutaneous renal allograft biopsies. Hum Pathol. 2001; 322205–215
81. Desvaux D, Le Gouvello S, Pastural M, et al. Acute renal allograft rejections with major interstitial oedema and plasma cell-rich infiltrates: high gamma-interferon expression and poor clinical outcome. Nephrol Dial Transplant. 2004; 194933–939
82. Gärtner V, Eigentler TK, Viebahn R. Plasma cell-rich rejection processes in renal transplantation: morphology and prognostic relevance. Transplantation. 2006; 817986–991
83. Rogers NM, Lawton PD, Jose MD. Plasma cell infiltrates and renal allograft outcomes in indigenous and non-indigenous people of the Northern Territory of Australia. Nephrology (Carlton). 2011; 168777–783
84. Gupta R, Sharma A, Mahanta PJ, et al. Plasma cell-rich acute rejection of the renal allograft: a distinctive morphologic form of acute rejection? Indian J Nephrol. 2012; 223184–188
85. Abbas K, Mubarak M, Zafar MN, et al. Plasma cell-rich acute rejections in living-related kidney transplantation: a clinicopathological study of 50 cases. Clin Transplant. 2015; 299835–841
86. Uppin MS, Gudithi S, Taduri G, et al. Expanding the antibody-mediated component of plasma cell-rich acute rejection: a case series. Indian J Nephrol. 2016; 263176–181
87. Dufek S, Khalil A, Mamode N, et al. Plasma-cell-rich infiltrates in paediatric renal transplant biopsies are associated with increased risk of renal allograft failure. Pediatr Nephrol. 2017; 324679–684
88. Hasegawa J, Honda K, Omoto K, et al. Clinical and pathological features of plasma cell-rich acute rejection after kidney transplantation. Transplantation. 2018; 1025853–859
89. Nishimura A, Masuzawa N, Nakamura T, et al. Clinicopathological and immunohistochemical analysis of plasma cell-rich rejection in renal transplantation: involvement of intratubular Th1/Th2 balance in plasma cell enrichment. Nephrology (Carlton). 2018; 23Suppl 2:52–57
90. Hamada AM, Yamamoto I, Kawabe M, et al. Clinicopathological features and outcomes of kidney allografts in plasma cell-rich acute rejection: a case series. Nephrology (Carlton). 2018; 23Suppl 2:22–26
91. Alhamoud I, Huang R, Lacelle C, et al. Allograft outcomes of treated children with kidney transplant who developed plasma cell-rich acute rejection (PCAR): a single center’s experience. Pediatr Transplant. 2019; 236e13500
92. Krejcik J, Casneuf T, Nijhof IS, et al. Daratumumab depletes CD38+ immune regulatory cells, promotes T-cell expansion, and skews T-cell repertoire in multiple myeloma. Blood. 2016; 1283384–394
93. Kwun J, Matignon M, Manook M, et al. Daratumumab in sensitized kidney transplantation: potentials and limitations of experimental and clinical use. J Am Soc Nephrol. 2019; 3071206–1219
94. Xu X, Shi B, Cai M, et al. A retrospective study of plasma cell infiltrates in explanted renal allografts. Transplant Proc. 2008; 4051366–1370
95. Aiello FB, Calabrese F, Rigotti P, et al. Acute rejection and graft survival in renal transplanted patients with viral diseases. Mod Pathol. 2004; 172189–196
96. Kemény E, Hirsch HH, Eller J, et al. Plasma cell infiltrates in polyomavirus nephropathy. Transpl Int. 2010; 234397–406
97. Trpkov K, Marcussen N, Rayner D, et al. Kidney allograft with a lymphocytic infiltrate: acute rejection, posttransplantation lymphoproliferative disorder, neither, or both entities? Am J Kidney Dis. 1997; 303449–454
98. Randhawa P, Demetris AJ, Pietrzak B, et al. Histopathology of renal posttransplant lymphoproliferation: comparison with rejection using the Banff schema. Am J Kidney Dis. 1996; 284578–584
99. Randhawa PS, Magnone M, Jordan M, et al. Renal allograft involvement by Epstein-Barr virus associated post-transplant lymphoproliferative disease. Am J Surg Pathol. 1996; 205563–571
100. Zimmermann H, Oschlies I, Fink S, et al. Plasmablastic posttransplant lymphoma: cytogenetic aberrations and lack of Epstein-Barr virus association linked with poor outcome in the prospective German Posttransplant Lymphoproliferative Disorder Registry. Transplantation. 2012; 935543–550
101. Trappe R, Zimmermann H, Fink S, et al. Plasmacytoma-like post-transplant lymphoproliferative disorder, a rare subtype of monomorphic B-cell post-transplant lymphoproliferation, is associated with a favorable outcome in localized as well as in advanced disease: a prospective analysis of 8 cases. Haematologica. 2011; 9671067–1071
102. Karuturi M, Shah N, Frank D, et al. Plasmacytic post-transplant lymphoproliferative disorder: a case series of nine patients. Transpl Int. 2013; 266616–622
103. Dharnidharka VR. Comprehensive review of post-organ transplant hematologic cancers. Am J Transplant. 2018; 183537–549
104. Pineda S, Sigdel TK, Liberto JM, et al. Characterizing pre-transplant and post-transplant kidney rejection risk by B cell immune repertoire sequencing. Nat Commun. 2019; 1011906
105. Cheng J, Torkamani A, Grover RK, et al. Ectopic B-cell clusters that infiltrate transplanted human kidneys are clonal. Proc Natl Acad Sci U S A. 2011; 108145560–5565
106. Dhir R, Nalesnik M, Demetris A, et al. Latent membrane protein expression in post-transplant lymphoproliferative disease. Appl Immunohistochem. 1995; 3:123–126
107. Abbas K, Mubarak M, Zafar MN, et al. Management of plasma cell-rich acute rejection in living-related kidney transplant: role of proteasome inhibitor. Exp Clin Transplant. 2019; 17142–46
108. Tasaki M, Saito K, Nakagawa Y, et al. Bortezomib eliminates plasma cells from a renal graft in plasma cell-rich acute rejection. Transplant Proc. 2019; 5161732–1738
109. Fillatreau S. Natural regulatory plasma cells. Curr Opin Immunol. 2018; 55:62–66
110. Baba Y, Saito Y, Kotetsu Y. Heterogeneous subsets of B-lineage regulatory cells (Breg cells). Int Immunol. 2019doi:10.1093/intimm/dxz068
Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.