Pitfalls in graft survival analysis.

The reliability of a scientific work depends on the accuracy of the analysis. Scientific publications in the field of kidney transplantation still contain methodical errors that may lead to wrong conclusions and result in severe consequences for the patients. Using the data from the Collaborative Transplant Study, we are presenting in this study six examples of the erroneous usage of statistical methods and show how these mistakes can be avoided.

Pre-transplant soluble CD30 in combination with total DSA but not pre-transplant C1q-DSA predicts antibody-mediated graft loss in presensitized high-risk kidney transplant recipients.

Presensitized kidney transplant recipients are at high-risk for early antibody-mediated rejection. We studied the impact of pre- and post-transplant donor-specific human leukocyte antigen (HLA) antibodies (DSA) and T-cell-activation on the occurrence of antibody-mediated rejection episodes (AMR) and graft loss (AMR-GL) in a unique cohort of 80 desensitized high-risk kidney transplant recipients. Patients with pre-transplant DSA demonstrated more AMR episodes than patients without DSA, but did not show a significantly increased rate of AMR-GL. The rates of AMR and AMR-GL were not significantly increased in patients with complement split product (C1q)-binding pre-transplant DSA. Pre-transplant C1q-DSA became undetectable post-transplant in 11 of 13 (85%) patients; 2 (18%) of these 11 patients showed AMR but no AMR-GL. In contrast, the post-transplant presence of C1q-DSA was associated with significantly higher rates of AMR (86 vs 33 vs 0%; P < 0.001) and AMR-GL (86 vs 0 vs 0%; log-rank P < 0.001) compared with post-transplant DSA without C1q-binding or the absence of DSA. Patients with both pre-transplant DSA and evidence of pre-transplant T-cell-activation as indicated by soluble CD30-positivity showed a significantly increased risk for AMR-GL [HR = 11.1, 95% confidence interval (CI) = 1.68-73.4; log-rank P = 0.013]. In these high-risk patients, AMR-GL was associated with total DSA in combination with T-cell-activation pre-transplant, and de novo or persistent C1q-binding DSA post-transplant.

Determination of unacceptable HLA antigen mismatches in kidney transplant recipients: recommendations of the German Society for Immunogenetics.

One of the major tasks of histocompatibility and immunogenetics laboratories is the pretransplant determination of unacceptable antigen mismatches (UAM) in kidney transplant recipients. In this procedure, human leucocyte antigen (HLA) specificities are defined against which the patient has circulating alloantibodies that are expected to harm the transplanted organ. Using the information on UAM and the potential donor's complete HLA typing, prediction of the crossmatch result, the so called 'virtual crossmatch', is possible. Currently, the laboratories are using different algorithms for the determination of UAM, and depending on the algorithm, more or fewer organ offers are excluded for patients with a similar antibody profile. In order to bring homogeneity into the allocation of organs to immunized patients in Germany, the German Society for Immunogenetics established, on the basis of current knowledge, recommendations for the determination of UAM. The UAM recommendations, which are thought to serve as a common tool for responsible physicians at different transplant centers, contain technical issues that need to be considered and are individualized for sensitized patients with a high or intermediate risk of antibody-mediated rejection. The present review contains these recommendations and puts them into perspective to current international practice.

Complement C5a receptor gene 450 C/T polymorphism in renal transplant recipients: association of the CT genotype with graft outcome.

Complement-mediated humoral rejection has become the main focus of research in organ transplantation. The aim of this study was to investigate the possible association of the complement C5aR gene 450 C/T polymorphism in antibody-mediated renal allograft rejection. This polymorphism was investigated in 290 first deceased donor kidney graft recipients with well functioning grafts and no rejection treatment during the first transplant year (WFG), 265 recipients with graft failure within the first transplant year (F), and 187 healthy controls. Frequency of the 450 CT genotype was lower in the total population of 555 kidney recipients (4.7%) than in 187 healthy controls (8.6%), but the difference was not statistically significant (P = 0.065). A significantly higher frequency of CT genotype was found in F patients (CT: 6.8%) when compared to WFG patients (CT: 2.8%, P = 0.027). The CT genotype was also significantly lower in WFG patients than in healthy controls (P = 0.009). Low frequency of the C5aR 450 CT genotype, which apparently is a feature of certain kidney diseases, appears to be associated with good graft outcome in kidney transplantation and might be helpful for identifying recipients who are at low risk for graft rejection.

Influence of test technique on sensitization status of patients on the kidney transplant waiting list.

The exquisitely sensitive single antigen bead (SAB) technique was shown to detect human leukocyte antigen (HLA) antibodies in sera of healthy male blood donors. Such false reactions can have an impact on critical decisions, especially with respect to the determination of unacceptable HLA-antigen mismatches in patients awaiting a kidney transplant. We tested pretransplant sera of 534 patients on the kidney waiting list using complement-dependent cytotoxicity (CDC), enzyme-linked immunosorbent assay (ELISA) and SAB in parallel. Evidence of HLA antibodies was obtained in 5% of patients using CDC, 14% using ELISA, and 81% using SAB. Among patients without history of an immunizing event, 77% showed evidence of HLA antibodies in SAB. In contrast 98% of these patients were negative in ELISA and CDC. In patients without an immunizing event, SAB-detected antibodies reacted not always weakly but with mean fluorescence intensity (MFI) values as high as 14 440. High-MFI-value antibodies were found in some of these patients with HLA specificities that are rather common in general population, consideration of which would lead to unjustified exclusion of potential kidney donors. False SAB reactions can be unveiled by testing with additional antibody assays. Denial of donor kidneys to recipients based on HLA-antibody specificities detected exclusively in the SAB assay is not advisable.

Superiority of AbCross enzyme-linked immunosorbent assay cross-match over the B-cell complement-dependent lymphocytotoxicity cross-match.

BACKGROUND: The AbCross enzyme-linked immunosorbent assay (ELISA) cross-match is a recently introduced solid phase cross-match technique with several technical advantages over the currently available Antibody Monitoring System ELISA cross-match. METHODS: In the present study, we investigated the potential superiority of AbCross over the traditional complement-dependent lymphocytotoxicity (CDC) B-cell cross-match (BXM). Pretransplant sera of 271 kidney transplant recipients who were transplanted at our center between 1998 and 2010 were tested in ELISA screening for the presence of human leukocyte antigen (HLA) antibodies and in AbCross and CDC for antibody reactivity against solubilized donor HLA class I and II antigens and donor B cells, respectively. RESULTS: Patients positive for HLA class I or II antibodies on ELISA screening had a significantly poorer graft outcome 2 years after transplantation than recipients who were negative for HLA antibodies (21% vs 6% graft loss; P = .002). Corresponding with this finding, 37 recipients positive for HLA antibodies in AbCross against donor HLA class I or II antigens had a 2-year post-transplant graft loss rate of 19%, which is significantly higher than the 8% rate in 186 recipients who were negative for both antibody classes in AbCross (P = .043). The 2-year graft loss rate in 34 AbCross positive but BXM negative patients was 21%, compared with 7% in 172 AbCross and BXM negative patients (P = .012) and 9% in 11 AbCross negative but BXM positive patients (P = .39). CONCLUSIONS: Our data indicate that the AbCross ELISA cross-match is superior to the CDC BXM, most likely because it detects antibodies against donor HLA antigens at a higher sensitivity.

Algorithms for the determination of unacceptable HLA antigen mismatches in kidney transplant recipients.

One of the major tasks of human leukocyte antigen (HLA) laboratories is the pretransplant determination of unacceptable HLA antigen mismatches (UAM) in organ transplant recipients. HLA antigen specificities are determined against which the patient has circulating alloantibodies that are expected to harm the transplanted organ. Using the information on UAM, negative crossmatch (XM) prediction or 'virtual XM' is possible when a potential donor's complete HLA typing is available. Before the introduction of solid-phase antibody detection assays, UAM were determined using the complement-dependent cytotoxicity methodology. After the introduction of the single antigen bead technique, however, various UAM determination algorithms have emerged. In this report, six different laboratories worldwide present how they determine UAM in their collective of kidney transplant recipients in the pretransplant phase and proceed thereafter to transplantation.

Recent developments in desensitization of crossmatch-positive kidney transplant recipients.

Currently, there are two major options for the successful and timely transplantation of sensitized kidney transplant recipients: (1) avoidance of the sensitization barrier using special allocation programs, or (2) desensitization. In the case of broadly sensitized kidney patients, a combination of both options might be necessary. This review focuses on new advances in desensitization of crossmatch-positive kidney transplant recipients which include immunoadsorption and the administration of new substances such as the complement C5 inhibitor eculizumab. Finally, integrated algorithms that combine different measures are acknowledged.

Kidney transplantation for high-risk sensitized patients - the "Heidelberg algorithm".

Presensitized patients on kidney transplant waiting lists have a lower chance of receiving a crossmatch-negative kidney and, if transplanted, are at increased risk of immunologic graft loss. We developed an algorithm for pretransplant risk estimation and peritransplant treatment which contains 7 independent steps: 1) pretransplant identification of high-risk patients; 2) good HLA match; 3) inclusion in the Acceptable Mismatch Program of Eurotransplant if applicable; 4 and 5) pre- and posttransplantation desensitization; 6) monitoring of donor-specific antibodies after transplantation; and 7) protocol biopsies. Between April 1, 2006, and March 31, 2009, 34 patients were transplanted according to this algorithm. Graft and patient survival rates did not differ from those in nonsensitized recipients. Antibody-mediated rejection episodes were <10% and infectious complications infrequent.

Current approaches to the management of highly sensitized kidney transplant patients.

With the introduction of sensitive antibody detection techniques, effective antibody elimination devices, therapeutic agents, such as bortezomib and eculizumab, and new concepts, such as Heidelberg algorithm, kidney paired exchange, and acceptable mismatch programs, several effective options are now available for the management of highly sensitized kidney transplant patients. However, as the number of human leukocyte antigen-mismatched transplantations is increasing with each year and as long as the elimination or long-term control of donor-specific antibody-producing clones remains an unresolved issue, sensitization will continue to represent a major problem in kidney transplantation.

Options for immunologic support of renal transplantation through the HLA and immunology laboratories.

HLA and immunology laboratories are an integral part of clinical kidney transplant programs. They assist transplant clinicians with evaluating the immunological suitability of potential recipients for transplantation and selecting donor-recipient combinations with a low risk of immunological failure. With sophisticated new techniques becoming available for posttransplant immunological monitoring, laboratories play an increasing supporting role during posttransplant follow up. The level of precision at which immunological testing predicts clinical outcome, however, leaves room for improvement. In this article, we summarize the current state of diagnostics, discuss problems, and point out promising developments.

Desensitization strategies enabling successful renal transplantation in highly sensitized patients.

Currently, the number of highly sensitized patients awaiting a renal transplant is increasing on the waiting lists of different organ exchange organizations. Due to the presence of antibodies against a broad variety of human leukocyte antigen (HLA) specificities, highly sensitized patients have a markedly reduced chance of receiving a crossmatch-negative organ. It has long been recognized that hyperacute rejection is associated with the presence of donor-specific anti-HLA antibodies at the time of transplantation. Meanwhile treatment protocols have been developed to achieve successful transplantation across antibody barriers. Therefore, the presence of donor-specific anti-HLA antibodies and a positive serological crossmatch are no longer considered as an absolute contraindication to renal transplantation. Mainly, two desensitization protocols have been established in order to overcome a positive crossmatch or to enhance the chance of highly sensitized patients to receive a crossmatch-negative organ: high-dose intravenous immunoglobulin (IVIg) or low-dose IVIg in combination with plasmapheresis. Herein, we summarize the characteristics of these two treatment regimes along with other alternative approaches that are currently used for the management of kidney graft recipients with broad alloantibody reactivity against potential kidney donors.

Post-transplant sCD30 and neopterin as predictors of chronic allograft nephropathy: impact of different immunosuppressive regimens.

Immunological monitoring for chronic allograft nephropathy (CAN) is of great potential interest. We assessed serum soluble CD30 (sCD30) together with in vitro Th2-type responses (IL-4, IL-10, CD4 helper activity) and neopterin in a prospective study of 84 renal transplant recipients with 2-year follow-up. Patients were randomized to CsA/Aza, CsA/MMF and Tacr/Aza, respectively, to analyze the effect of immunosuppression on posttransplant sCD30 and neopterin. ATG induction and acute rejections did not alter sCD30 levels whereas CMV disease was associated with transient upregulation of sCD30 (p = 0.003 at 4 months) and sustained upregulation of neopterin (corrected for graft function (Neo/CR) p = 0.005 at 2 years). Tacr versus CsA treatment proved to be an independent variable associated with downregulation of 1-year sCD30, which was positively related to Neo/CR (p = 0.007 and 0.01, respectively; logistic regression). Importantly, increased 1-year sCD30 and Neo/CR were associated with decreased glomerular filtration rate at 2 years (p = 0.02 and p < 0.0005, respectively) and evidence of CAN (p < 0.0005). High 1-year sCD30 could not be attributed to enhanced Th2-type responses and was not associated with HLA antibody formation. Our data suggest that elevated sCD30 and neopterin predict graft deterioration by CAN. Tacr effectively downregulates these responses and might be of advantage in patients with elevated sCD30 or neopterin.

sCD30 and neopterin as risk factors of chronic renal transplant rejection: impact of cyclosporine A, tacrolimus, and mycophenolate mofetil.

High pretransplantation sCD30 levels have been shown to be associated with lower 5-year kidney graft survival in mainly Cyclosporine A (CsA)-treated recipients (Collaborative Transplant Study database). To analyze the effect of different immunosupressive regimens (CsA/Azathioprine [Aza], CsA/Mycophenolate Mofetil [MMF], Tacrolimus [Tacr]/Aza) on sCD30, we assessed serum sCD30 and neopterin together with in vitro cytokine responses in a prospective randomized study of 84 renal transplant recipients before, 4 months, and 1 year after transplantation. Panel-reactive antibody (PRA) formation, HLA matching, ATG induction therapy, and acute rejections had no impact on sCD30 levels, whereas cytomegalovirus (CMV) infections induced an up-regulation of sCD30 4 months posttransplantation (P = .003). Whereas MMF showed no effect on sCD30 compared with Aza therapy, we found a significant impact of Tacr versus CsA treatment (1-year sCD30 > or = 60 U/mL: 14/42 (33%), CsA; 1/38 (3%), Tacr; P < .0005). Chronic rejection 2 years posttransplantation was associated with elevated 1-year sCD30 (P = .001) and neopterin levels (P = .006). Our data indicate that the Th2 activation marker sCD30 provides a risk factor for chronic rejection independent of classical immunological risk factors and may be down-regulated using Tacr treatment.

Advances in pre- and posttransplant immunologic testing in kidney transplantation.

Pre- and posttransplant risk estimation in kidney transplantation is important for the selection of appropriate treatment strategies. Recently, using new immunologic tests, we made observations within the framework of the Collaborative Transplant Study that may influence clinical practice. Complement-dependent lymphocytotoxic panel reactivity as a measure of anti-HLA sensitization, although criticized for its low sensitivity, is a useful indicator of an increased risk of rejection. Using the more sensitive complement-independent ELISA methodology, which utilizes solubilized HLA molecules instead of lymphocytes, we found that recipients with preformed complement-dependent anti-HLA antibodies showed a decreased graft survival only if their antibodies were directed against both HLA class I and class II, whereas isolated reactivity only against HLA class I or class II was of no clinical consequence. Pretransplant serum-soluble CD30 (sCD30) was found to be an independent and highly predictive factor of immunologic risk. The effects of sCD30 and anti-HLA antibodies were additive. Importantly, even patients without anti-HLA antibodies showed a strong HLA matching effect if their pretransplant serum contained high levels of sCD30. Although the role of anti-HLA antibody formation after transplantation remains uncertain, ELISA-detected sCD30 was shown to indicate impending graft rejection as early as on posttransplant days 3 to 5. Another very sensitive indicator of impending rejection is provided by posttransplant monitoring of the cytotoxic T-lymphocyte effector genes, perforin and granzyme B, in peripheral blood using real-time PCR.

Toxic effects of the herbicide 2,4-dichlorophenoxyacetic acid on lymphoid organs of the rat.

We investigated the toxicity of the widely used herbicide dimethylammonium salt of 2,4-dichlorophenoxyacetic acid (DMA-2,4-D) on the lymphoid system of rats after a single dose oral administration using histological, cytochemical and molecular methods. DMA-2,4-D destroyed in a dose- and time-dependent manner the vascular integrity of the thymus and caused cell depletion in the white pulp of the spleen and in the cortex of the thymus, which was at least partly due to programmed cell death. DMA-2,4-D appeared to have hemolytic activity and caused intravascular hemolysis of erythrocytes. An increased hemosiderin content in macrophages of the spleen indicated phagocytosis of lysed erythrocytes. We conclude that acute DMA-2,4-D herbicide poisoning results in severe damage of the lymphatic organs thymus and spleen.

Increased soluble Fas in HIV-infected hemophilia patients with CD4+ and CD8+ cell count increases and viral load and immune complex decreases.

Previous studies interpreted increases of soluble Fas (sFas) in the plasma during disease progression in HIV-infected patients as evidence of increased apoptosis of CD4(+) lymphocytes. We studied whether sFas and sFas ligand (sFasL) plasma levels are associated with CD4(+) and CD8(+) lymphocyte counts, plasma viral load, and IgM, IgG, C3d, and gp120 complexes on circulating CD4(+) blood lymphocytes in long-term surviving HIV-infected hemophilia patients, most of whom were receiving HAART. Twenty-six hemophilia patients who were infected with HIV in the early 1980s were investigated in 1997, 1998, and 1999. HAART was initiated in 1996 and 1997 in most patients. Lymphocyte subpopulations and immune complex-coated CD4(+) lymphocytes in the blood were investigated by flow cytometry, plasma viral load (HIV-1 mRNA copies/ml plasma) was tested with HIV-1 QT Nuclisens kits, sFas (ng/ml) and sFasL (ng/ml) plasma levels were measured with MBL ELISA kits, and the in vitro response of patient lymphocytes was tested in cell cultures. During the period from 1997 to 1999 we observed an increase in sFas plasma levels (p = 0.003) as well as in CD4(+) (p = 0.004) and CD8(+) (p = 0.023) cell counts; a decrease in IgG (p = 0.047), C3d (p = 0.024), and gp120 (p = 0.001)-coated CD4(+) lymphocytes in the blood; and a decrease in the number of impaired mitogen stimulation assays (p = 0.013). sFas was negatively associated with viral burden (r = -0.662, p = 0.0002) as well as with CD4(+)IgM(+) (r = -0.554, p = 0.004), CD4(+)IgG(+) (r = -0.431, p = 0.031), CD4(+)C3d(+) (r = -0.551, p = 0.041), and CD4(+)gp120(+) (r = -0.430, p = 0.041) blood lymphocytes, CD8(+)DR(+) cell counts (r = -0.700, p = 0.016), and impaired in vitro responses of patient lymphocytes to PHA (r = -0.475, p = 0.016). sFasL was negatively associated with total lymphocyte counts (r = -0.433, p = 0.027), as well as with absolute numbers of CD3(+) (r = -0.492, p = 0.011) and CD8(+) (r = -0.432, p = 0.027) cells. We conclude that, contrary to expectations, sFas plasma levels increased in long-term surviving HIV-infected hemophilia patients receiving HAART, concomitant with increases in CD4(+) and CD8(+) cell counts. Increased sFas may reflect the growing pool of T lymphocytes that recovers because of a decreasing viral burden and a decreasing immune complex load of CD4(+) lymphocytes.

Association of immune complexes and plasma viral load with CD4+ cell depletion, CD8+ DR+ and CD16+ cell counts in HIV+ hemophilia patients. Implications for the immunopathogenesis of HIV-induced CD4+ lymphocyte depletion.

OBJECTIVE: There is evidence that HIV induces CD4+ depletion in part by the formation of immune complexes (IC) that attach to CD4+ blood lymphocytes. In the present study we examined the relationship of IC-coated CD4+ blood cells with retroviral replication in HAART-treated patients. PATIENTS AND METHODS: 52 hemophilia patients were studied from 1997 to 1999. Lymphocyte subsets, IgM, IgG and gp120 on CD4+ blood cells, in vitro responses of lymphocytes to mitogens, plasma neopterin and plasma viral load were measured. RESULTS: Patients with detectable viral replication and without ICs on CD4+ blood lymphocytes had a lower viral load (4100 versus 21000 HIV-1 mRNA copies/ml; P = 0.079) and higher CD4+ cell counts (310/microl versus 161/microl; P = 0.035) than patients with ICs on circulating CD4+ lymphocytes. Among patients with < 80 HIV-1 mRNA copies/ml, IC- individuals had slightly higher CD4+ lymphocyte counts than IC+ patients (384/microl versus 316/microl; n.s.). Further evidence for the clinical relevance of the ICs was obtained when 18 patients who had an undetectable viral load at previous investigations were analyzed. Among patients with a stable undetectable viral load, CD4+ counts increased in 6 of 8 IC- but in none of 2 IC+ individuals. In patients whose viral load increased during the observation period, 5 of 6 IC- but none of 2 IC+ individuals showed higher CD4+ cell counts. Impaired virus killing is suggested by lower CD16+ (35/microl versus 107/microl; P = 0.016), higher CD3+ DR+ (178/microl versus 66/microl; P = 0.006), and higher CD8+ DR+ (142/microl versus 34/microl; P = 0.017) cell counts in IC(-) patients compared to IC- patients without detectable viral load. Strong retroviral replication induced strong T cell dysfunctions. Fewer CD3+ 25+ blood lymphocytes (19/microl versus 47/microl; P = 0.006) and a lower in vitro response of T lymphocytes to the mitogens Con A (RR: 0.3 versus 1.2; P=0.023) and CD3 mab (RR: 0.5 versus 2.4; P = 0.012) was observed in IC+ patients with detectable versus undetectable viral load. CONCLUSION: Our data suggest that ICs on circulating CD4+ blood lymphocytes are primarily associated with CD4+ lymphocyte depletion whereas the plasma viral load is primarily associated with decreased T lymphocyte activation, lower CD16+ counts, and higher CD8+ DR+ lymphocytes which might be the effector cells for virus elimination.

Induction of apoptosis in human lymphocytes by the herbicide 2,4-dichlorophenoxyacetic acid.

Dimethylammonium salt of 2,4-dichlorophenoxyacetic acid (DMA-2,4-D) is a widely used herbicide that is considered moderately toxic. In the present study we found that DMA-2,4-D is able to cause apoptosis in peripheral blood lymphocytes of healthy individuals and Jurkat T cells. Apoptosis induced by DMA-2,4-D was dose and time dependent, independent of Fas, TNF receptor 1 or the aromatic hydrocarbon receptor, and involved disruption of the mitochondrial transmembrane potential and activation of caspase-9. ZVAD-FMK, a broad-spectrum inhibitor of caspases, blocked DMA-2,4-D-induced apoptosis completely. While an inhibitor of caspase-9, as well as caspase-9 and caspase-3 inhibitors in combination, strongly blocked DMA-2,4-D-induced apoptosis, an inhibitor of caspase-3 had a moderate inhibitory effect. Unlike Fas-mediated apoptosis, the initiator caspase, caspase-8, was not involved in DMA-2,4-D-induced apoptosis. Transfection of Jurkat cells with Bcl-2 prevented DMA-2,4-D-induced disruption of the mitochondrial transmembrane potential and led to a complete blockage of apoptosis. Our data indicate that DMA-2,4-D kills human lymphocytes by initiating apoptosis via a direct effect on mitochondria. The activation of caspases occurs downstream of mitochondrial damage, and the dysfunction of mitochondria appears to be sufficient for triggering all downstream events leading to apoptosis.