Shili Ge

Infectious Complications in Kidney-Transplant Recipients Desensitized with Rituximab and Intravenous Immunoglobulin

BACKGROUND AND OBJECTIVES Rituximab and intravenous Ig (IVIG) are commonly used for desensitization of HLA and blood group-incompatible (ABOi) transplants. However, serious infections have been noted in association with rituximab administration. In this study, we retrospectively compared infectious outcomes in those who received rituximab plus IVIG for HLA or ABOi transplants (RIT group) with a group of nonsensitized, ABO-compatible transplant recipients (non-RIT group). DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS Patients undergoing kidney transplantation at Cedars-Sinai Medical Center were included in the analysis. A total of 361 patients were identified. All received antimicrobial prophylaxis and viral surveillance. The primary outcome was infection. RESULTS Overall patient survival was 97 and 96%, and graft survival was 91 and 89% in the RIT and non-RIT groups, respectively, after an average follow-up of 18 months. There were equal rates of bacterial (34.7% versus 39.1%), viral (21.8% versus 25.1%), fungal (5.9% versus 5.2%), and serious infections (22.9% versus 25.5%) in the RIT and non-RIT groups respectively. Urinary tract infection was the most common infection, accounting for 50% of all bacterial infections. Cytomegalovirus viremia was nonsignificantly more common in the nonrituximab-treated group (15.2% versus 10%), whereas BK viremia was marginally more frequent in the rituximab-treated group (10.6% versus 5.8%). There were no graft losses caused by BK-associated nephropathy. There were two deaths in each group related to infection (1%). CONCLUSION Rituximab does not increase infection risk when used with intravenous Ig for desensitization.

Benefits of rituximab combined with intravenous immunoglobulin for desensitization in kidney transplant recipients.

Background Highly HLA-sensitized (HS) patients have difficulty accessing compatible donors, especially deceased donor (DD) transplants. Desensitization protocols (DES) have evolved, but rigorous evaluation is lacking. Here, we examined the efficacy of rituximab as a DES agent in a placebo-controlled trial. Methods Candidates were randomized to IVIG+placebo versus IVIG+rituximab. End points included rates of transplantation, antibody-mediated rejection (ABMR), and renal function. Protocol biopsies were performed at 1 year and analysis of patient and graft survival and donor-specific HLA antibodies (DSA) were performed. Results Initially, 15 HS DDs were randomized with 13 receiving transplants. However, we discontinued study entry after five serious adverse events were observed. The study was un-blinded and attribution of patients was noted (IVIG+placebo N=7, IVIG+rituximab N=6). No significant differences were seen in DSA levels at transplant. All ABMR episodes occurred in the IVIG+placebo arm and required intense therapy (P=0.06). The two graft losses were in the placebo group. DSA rebound associated with severe ABMR was seen in three patients in the IVIG+placebo group. No rebound was seen in the IVIG+rituximab group. Renal function at 6 and 12 months showed a significant benefit for IVIG+rituximab (P=0.04). Conclusions Based on limited assessment with acknowledged limitations, both protocols appear effective in achieving levels of DSA allowable for transplantation. However, IVIG+rituximab appeared more effective in preventing DSA rebound and, more importantly, preventing ABMR and development of transplant glomerulopathy.

A Phase I/II Trial of the Interleukin-6 Receptor-Specific Humanized Monoclonal (Tocilizumab) + Intravenous Immunoglobulin in Difficult to Desensitize Patients.

Background Current desensitization (DES) methods are not always effective. Thus, novel, more effective approaches are desirable. Interleukin (IL)-6 is an attractive target as it promotes B-cell differentiation to plasma cells, is important for immunoglobulin production, and induces Th17 cells. Here, we undertook a phase I/II pilot study of DES using a novel drug (anti–IL-6 receptor (IL-6R),Tocilizumab [TCZ]) + intravenous Ig (IVIg) to assess safety and limited efficacy. Methods From July 2012 to November 2013, 10 patients unresponsive to DES with IVIg + Rituximab were treated with IVIg + TCZ. Patients received IVIg on days 0 and 30 at 2 g/kg and TCZ 8 mg/kg on day 15 then monthly for 6 months. If transplanted, patients received IVIg once and TCZ monthly for 6 months. Results No differences in baseline characteristics were seen in patients not transplanted versus transplanted. Two patients in each group developed serious adverse events: not transplanted- pulmonary congestion with epilepticus (likely not related) versus transplanted infective colitis with colonic perforation and Bell Palsy (both possibly related). Five of 10 patients were transplanted. Mean time to transplant from first DES was 25 ± 10.5 months but after TCZ was 8.1 ± 5.4 months. Six-month protocol biopsies showed no antibody-mediated rejection. Donor-specific antibody strength and number were reduced by TCZ treatment. Renal function at 12 months was 60 ± 25 mL/min. Conclusions Tocilizumab and IVIg appear to be safe. From this pilot trial, we are cautiously optimistic that targeting the IL-6/IL-6R pathway could offer a novel alternative for difficult to desensitize patients. Larger controlled studies are essential to prove efficacy.

IgG Endopeptidase in Highly Sensitized Patients Undergoing Transplantation

Background Donor‐specific antibodies create an immunologic barrier to transplantation. Current therapies to modify donor‐specific antibodies are limited and ineffective in the most highly HLA‐sensitized patients. The IgG‐degrading enzyme derived from Streptococcus pyogenes (IdeS), an endopeptidase, cleaves human IgG into F(ab′)2 and Fc fragments inhibiting complement‐dependent cytotoxicity and antibody‐dependent cellular cytotoxicity, which suggests that IdeS might be useful for desensitization. We report on the combined experience of two independently performed open‐label, phase 1–2 trials (conducted in Sweden and the United States) that assessed the efficacy of IdeS with regard to desensitization and transplantation of a kidney from an HLA‐incompatible donor. Methods We administered IdeS to 25 highly HLA‐sensitized patients (11 patients in Uppsala or Stockholm, Sweden, and 14 in Los Angeles) before the transplantation of a kidney from an HLA‐incompatible donor. Frequent monitoring for adverse events, outcomes, donor‐specific antibodies, and renal function was performed, as were renal biopsies. Immunosuppression after transplantation consisted of tacrolimus, mycophenolate mofetil, and glucocorticoids. Patients in the U.S. study also received intravenous immune globulin and rituximab after transplantation to prevent antibody rebound. Results Recipients in the U.S. study had a significantly longer cold ischemia time (the time elapsed between procurement of the organ and transplantation), a significantly higher rate of delayed graft function, and significantly higher levels of class I donor‐specific antibodies than those in the Swedish study. A total of 38 serious adverse events occurred in 15 patients (5 events were adjudicated as being possibly related to IdeS). At transplantation, total IgG and HLA antibodies were eliminated. A total of 24 of 25 patients had perfusion of allografts after transplantation. Antibody‐mediated rejection occurred in 10 patients (7 patients in the U.S. study and 3 in the Swedish study) at 2 weeks to 5 months after transplantation; all these patients had a response to treatment. One graft loss, mediated by non‐HLA IgM and IgA antibodies, occurred. Conclusions IdeS reduced or eliminated donor‐specific antibodies and permitted HLA‐ incompatible transplantation in 24 of 25 patients. (Funded by Hansa Medical; ClinicalTrials.gov numbers, NCT02224820, NCT02426684, and NCT02475551.)

Regulation of anti-HLA antibody-dependent natural killer cell activation by immunosuppressive agents.

Background It was demonstrated that human natural killer (NK) cells, via antibody-dependent cellular cytotoxicity (ADCC)-like mechanism, increase IFN&ggr; production after exposure to alloantigens. This finding was associated with an increased risk for antibody-mediated rejection (ABMR). Although the effects of various immunosuppressive drugs on T cells and B cells have been extensively studied, their effects on NK cells are less clear. This study reports the effect of immunosuppressive agents on antibody-mediated NK cell activation in vitro. Methods Whole blood from normal individuals was incubated with irradiated peripheral blood mononuclear cells (PBMCs) pretreated with anti-HLA antibody+ sera (in vitro ADCC), with or without immunosuppressive agents. The %IFN&ggr;+ and CD107a+ (degranulation marker) in CD56+ NK cells were enumerated by flow cytometry. Results Cyclosporine A and tacrolimus significantly reduced IFN&ggr; production in a dose-dependent manner (53%–83%), but showed minimal effect on degranulation (20%). Prednisone significantly reduced both IFN&ggr; production and degranulation (50%–66% reduction at maximum therapeutic levels). Calcineurin inhibitors (CNIs) in combination with prednisone additively suppressed IFN&ggr; production and degranulation. The effect of sirolimus or mycophenolate mofetil on NK cells was minimal. Conclusions These results suggest that potent suppressive effects of CNIs and prednisone on antibody-mediated NK cell activation may contribute to the reduction of ADCC in sensitized patients and possibly reduce the risk for ADCC-mediated ABMR. These further underscore the importance of medication compliance in prevention of ABMR and possibly chronic rejection, and suggest that ADCC-mediated injury may increase in strategies aimed at CNI or steroid minimization or avoidance.

IFNγ production by NK cells from HLA-sensitized patients after in vitro exposure to allo-antigens

Using a novel cytokine flow cytometry test (allo-CFC), we have previously shown that incubation of allogeneic cells with peripheral blood from highly-HLA sensitized (HS) patients results in reproducible gamma-interferon (IFNγ production in CD3(-) cells, and high (+) allo-CFC levels correlated with risk for antibody-mediated rejection (AMR). Here we report on identification of the cells and mechanisms responsible. The allo-CFC with/without modification was performed using blood from HS or normal individuals. IFNγ producing cells were CD3(-)/CD19(-), but CD3(-)/CD56(+). In vitro and in vivo B cell-depletion did not affect IFNγ production, demonstrating NK cells as the cells responsible for IFNγ production. NK cells from allo-CFC(+) or (-) individuals released significant amounts of IFNγ against target cells treated with serum from allo-CFC(+) individuals, but not allo-CFC(-) individuals. IFNγ release was abrogated by protein A/G treatment of the pretreated target cells, suggesting mediation by antibodies via FcγRIIIa (CD16). In conclusion, NK cell IFNγ release after allo-antigen exposure is mediated primarily through antibody-dependent cellular cytotoxicity (ADCC)-like mechanisms, suggesting that NK cells may be partially responsible for graft injury during AMR including C4d(-) AMR via ADCC, and could be a potential target for modification of this process.

Immunologic parameters and viral infections in patients desensitized with intravenous immunoglobulin and rituximab

Desensitization with IVIG and rituximab followed by transplantation with alemtuzumab or daclizumab induction is an effective clinical protocol. Here, we examined the effects of this protocol on immune cell number, T cell function by Cylex ImmuKnow®, CMV-specific CD8+ T cell (CMV-Tc) activity, total and viral-specific immunoglobulin levels and viral infections. In 17 highly HLA-sensitized (HS) patients who received desensitization, CD19+ cells were undetectable immediately after desensitization, while other immune cells were unchanged. No alteration in Cylex or CMV-Tc levels was seen. In separate 14 HS patients who were desensitized followed by transplantation, T cell numbers were near zero after alemtuzumab, while NK cell reduction was minimal. Early B cell recovery was not a risk for antibody-mediated rejection. Total IgG, IgM, and IgA remained in the normal range up to 12.6 months post-transplant, and CMV IgG level did not change. CMV-Tc activity was eliminated post-transplant in some patients, but recovered by 4 months post-transplant. None of them developed CMV infection. In conclusion, IVIG-rituximab-desensitization does not significantly alter T cell function pre-transplant, or reduce Ig levels below the normal range post-transplant. Although post-transplant induction therapy is associated with a transient depletion of viral-specific CD8+ memory cells, it does not increase risks for viral infections.

Intracellular IFNγ production in CD3 negative cells exposed to allo-antigens is an indicator of prior sensitization

BACKGROUND Sensitization to HLA antigens (Ags) is a significant obstacle to kidney transplantation and risk factor for antibody-mediated rejection (AMR). Current screening methods to assess HLA Ag exposure include various antibody assays. However, tools to accurately measure cell-mediated immunity to allo-Ags in a clinical setting are lacking. Here we report on an intracellular cytokine flow cytometry (CFC) assay that detects intracellular gamma-interferon (IFNgamma) production in non-T cell populations (CD3-) that appears to assess sensitization from previous allo-Ag exposure. METHODS Blood from 106 highly-HLA sensitized (HS) patients (pre-, post-IVIG-treatment [Rx] and/or post-transplant) and 14 3(rd) party normal controls (3(rd)N) were incubated with donor or 3(rd)N peripheral blood mononuclear cells (PBMCs), and IFNgamma+/CD3- cells were enumerated. RESULTS The percentage of IFNgamma+ cells in CD3- cells without stimulation in pre-IVIG-Rx HS patients was similar to normals, but significantly increased with incubation with donor and/or 3(rd)N PBMCs. Reactivity in normals was minimal. Reactivity was higher in HS females than HS males. Normal females with previous pregnancy (PG) showed significantly higher response than females without PG or non-sensitized normal males. Donor-specific reactivity in the CFC assay better correlated with donor-specific B cell crossmatch than total anti-HLA antibody levels or PRA. HS patients who developed AMR post-transplant showed significantly higher reactivity than those without AMR. CONCLUSIONS The CFC assay measures IFNgamma production in CD3- cells that may indicate a memory response to allo-Ags. This response is limited to HS patients and normal females with previous PG. Patients undergoing AMR show significantly higher reactivity. This assay may represent a novel approach to measurement of allo-sensitization with clinical utility in predicting those at risk for AMR.

JC polyomavirus viremia and progressive multifocal leukoencephalopathy in human leukocyte antigen-sensitized kidney transplant recipients desensitized with intravenous immunoglobulin and rituximab

Desensitization (DES) with intravenous immunoglobulin (IVIG) + rituximab is effective, safe, and increases the transplantation rate in human leukocyte antigen‐sensitized patients. However, reports of progressive multifocal leukoencephalopathy (PML) caused by JC polyomavirus (JCPyV) in autoimmune patients treated with rituximab is concerning. Here, we report on the JCPyV viremia and PML status in kidney transplant patients with/without DES (non‐DES).

Genes associated with antibody-dependent cell activation are overexpressed in renal biopsies from patients with antibody-mediated rejection

INTRODUCTION Antibody-mediated rejection (ABMR) is dependent on complement activating donor-specific anti-HLA antibodies (DSA). This is commonly detected by C4d deposition in allografts. However, recent data define a C4d negative ABMR phenotype suggesting a role for complement-independent DSA injury, antibody-dependent cellular cytotoxicity (ADCC). METHODS Here, we established an in vitro ADCC model that identified human ADCC-activated genes using microarray analysis. We subsequently interrogated renal allograft biopsies from patients with ABMR and controls for mRNA expression of the ADCC-activated gene set. RESULTS We identified 13 ADCC-activated genes. Six gene expression assays including 8 of the 13 genes (CCL3, CCL4/CCL4L1/CCL4L2, CD160, IFNG, NR4A3 and XCL1/XCL2) were analyzed in 127 kidney biopsies obtained from HLA-sensitized (HS), non-HS patients and control individuals. Most ADCC-activated genes showed significantly higher expression in the transplant samples compared to the controls (p<0.0005). The gene expression levels were significantly higher in HS and non-HS transplant patients who developed ABMR compared to those who did not (p=0.04-0.002). There was no difference in the gene expression levels between C4d positive and negative ABMR (p=0.26-0.99). Samples from high PRA (>80%) or positive DSA patients showed higher gene expression levels for the ADCC-activated genes compared to low PRA (<80%) and negative DSA patients (p=0.04-0.001). CONCLUSION ADCC pathways are active in transplant patients with ABMR, and likely mediate allograft injury, providing a potential mechanism for C4d negative ABMR.