Lack of association between thrombosis-associated and cytokine candidate gene polymorphisms and acute rejection or vascular complications after kidney transplantation

Nephrology Dialysis Transplantation, Jan 2008

Background. Acute rejection episodes and vascular complications are common after renal transplantation and have negative impact on the long-term patient and graft survival. We investigated whether the risks of acute rejection, thrombosis, infarction and graft loss could be predicted based on the presence of functional polymorphisms in the genes of the coagulation and endothelial inflammation cascade. Methods. The study consisted of 772 consecutive cadaver kidney transplantations from a single centre. The effects of gene polymorphisms FVL, F5R2, FII G20210A, MTHFR C677T, F13A1 V34L, TFPI P151L, PROC W380G, TNF G(-308)A, IL10 A(-592)C, IL10 A(-1082)G and IL6 C(-174)G of recipients and donors were investigated. Results. We were unable to find statistically significant associations between any of the studied polymorphisms and clinical outcomes. Conclusions. Our results indicate that high-risk renal transplant candidates cannot be identified through the routine analysis of the polymorphisms.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://ndt.oxfordjournals.org/content/23/1/364.full.pdf

Lack of association between thrombosis-associated and cytokine candidate gene polymorphisms and acute rejection or vascular complications after kidney transplantation

Noora S. Alakulppi 1 2 Lauri E. Kyllo nen 0 1 Jukka Partanen 1 2 Kaija T. Salmela 0 1 Jarmo T. Laine 1 2 0 Renal Transplant Unit, Helsinki University Hospital , Kasarminkatu 11-13, FI-00130 Helsinki, Finland 1 Development, Finnish Red Cross Blood Service , FI-00310 Helsinki 2 Research and Development, Finnish Red Cross Blood Service , Kivihaantie 7 Background. Acute rejection episodes and vascular complications are common after renal transplantation and have negative impact on the long-term patient and graft survival. We investigated whether the risks of acute rejection, thrombosis, infarction and graft loss could be predicted based on the presence of functional polymorphisms in the genes of the coagulation and endothelial inflammation cascade. Methods. The study consisted of 772 consecutive cadaver kidney transplantations from a single centre. The effects of gene polymorphisms FVL, F5R2, FII G20210A, MTHFR C677T, F13A1 V34L, TFPI P151L, PROC W380G, TNF G(-308)A, IL10 A(-592)C, IL10 A(-1082)G and IL6 C(-174)G of recipients and donors were investigated. Results. We were unable to find statistically significant associations between any of the studied polymorphisms and clinical outcomes. Conclusions. Our results indicate that high-risk renal transplant candidates cannot be identified through the routine analysis of the polymorphisms. Introduction After kidney transplantation, long-term patient and graft survival are influenced by several complications. These include acute rejection and thrombotic episodes e.g. graft and deep vein thrombosis, myocardial and cerebral infarctions. It has been suggested that the development of acute rejection, thrombosis and infarction may share at least parts of the same pathophysiological pathway involving genes regulating the immune system of the vascular endothelium and those influencing thrombogenesis [1]. Up to 4-fold increases in the incidence of thromboembolic episodes has been reported in patients with pro-thrombotic gene polymorphisms [2]. Should polymorphisms affecting the expression levels of the genes have a major effect on the likelihood of the patient developing complications, patients with increased risk could be identified by determining the presence of the polymorphisms before transplantation. The immunosuppressive and antithrombotic medication of high-risk individuals could subsequently be tailored to diminish the likelihood of adverse outcomes. Previously, it has been reported that polymorphisms in the genes coagulation factor V (F5), coagulation factor II (F2), 5,10-methylenetetrahydrofolate reductase (MTHFR), tumor necrosis factor (TNF), interleukin 10 (IL10) and interleukin 6 (IL6) have a role in the development of acute rejection and vascular complications after renal transplantation [28]. In addition, polymorphisms in the genes coding for protein C (PROC), tissue factor pathway inhibitor (TFPI) and coagulation factor XIII A1 polypeptide (F13A1) have been previously found to either increase (PROC, TFPI) or decrease (F13A1) the risk of thrombosis in the general patient population [911]. We investigated whether the genetic variation in the candidate genes known to have a functional role in thrombogenesis (FVL, F5R2, FII G20210A, MTHFR C677T, F13A1 V34L, TFPI P151L, PROC W380G) or cytokine-mediated inflammation of vascular endothelium (TNF G(-308)A, IL10 A(-592)C, IL10 A(-1082)G and IL6 C(-174)G) would allow the identification of patients at a high risk of developing acute rejection episodes or vascular complications after kidney transplantation. The immunosuppressive and antithrombotic medication of high-risk individuals could subsequently be tailored to diminish the likelihood of adverse outcomes. In addition to recipient polymorphisms, we investigated the effects of donor polymorphisms. According to the vascular bed-specific hypothesis of thrombosis, genetic factors of the donor may, in an organ-specific fashion, influence graft outcome [12]. In particular, donor polymorphisms in genes regulating inflammation and coagulation may result in the donor kidney endothelium reacting inappropriately to recipient-derived stimuli [13]. Whether the donor kidney has an influence on thrombotic cascades in other organs than the kidney by secreting molecules, the gene expression level of which are regulated by donor gene polymorphisms, is yet to be studied. The present investigation was based on single-centre material including the data of adult cadaver kidney transplantations performed during five consecutive years (19992003) [14]. Further, genetic variation in the Finnish population is known to be relatively low due to founder effect and isolation, decreasing the likelihood of genetic heterogeneity influencing study results. Subjects and methods Patients The study population comprised 772 adult cadaver kidney transplantations for 764 recipients (eight recipients had two transplantations during the study period) and their 462 cadaver donors transplanted between 1999 and 2003 at the Renal Transplant Unit of Helsinki University Hospital, Finland (Table 1). Transplantations from living donors were excluded from this study. Median follow-up was 1494 days (minmax 12659 days). A total of 636 recipients received triple immunosuppression cyclosporine/tacrolimus, azathioprine/mycophenolate mofetil and methylprednisone and 136 received additional antibody induction therapy (anti-thymocyte globulin, basiliximab or daclizumab). An acute rejection episode was defined based on clinical and biopsy findings (Banff criteria). Delayed graft function (DGF) was defined as serum creatinine >500 mmol/l throughout the first week, >1 dialysis sessions, or oliguria >2 days. Data regarding vascular complications were retrospectively obtained from patient records. The study protocol was approved by the Ethical Review Board of the Helsinki University Hospital, Helsinki, Finland in accordance with the ethical standards laid down in an appropriate version of the 2000 Declaration of Helsinki. Samples Whole blood samples were originally taken prior to transplantation for histocompatibility testing. The extracted DNA or buffy coats from citrate anti-coagulated peripheral blood were stored at 208C. DNA was extracted using the QIAamp DNA Blood Mini Kit or the FlexiGene DNA Kit (QIAGEN Inc, CA, USA). For HLA-A and HLA-B typing, a serological lymphocytotoxicity method and for HLA-DRB1 typing a PCR-SSO (sequence specific oligonucleotides) method was used. For HLA class I antibody, screening flowcytometric PRA I Table 1. Overall characteristics of recipients at transplantation (n 772) and cadaver donors (none living) (n 462) Variable n (%) Age (years x) (min-max) 49.2 (17.8 74.5) 50 years 396 (51.3) < 50 years 376 (48.7) Recipient diagnosis Type I diabetes 167 (21.6) Type II diabetes 27 (3.5) Polycystic kidney 160 (20.7) Glomerulonephritis 194 (25.1) Other 224 (29.0) Waiting time (days x) (min-max) 407 (1 6682) 1st/2nd/3rd/4th allograft 687/73/11/1 (89.0/9.5/1.4/0.1) 136 (17.6) 210 (27.2) 2.2 (0 4) NA NA, not available/applicable; DGF, delayed graft function; MM, mismatch; PRA last, panel reactive antibody before transplantation; CIT, cold ischaemia time. (One Lambda, USA) and for pre-transplant cross matches a lymphocytotoxicity method was used. Thrombosis-associated gene polymorphism typing Recipients and donors were genotyped for F5 G1691A rs6025, F5 A4070G rs1800595, F2 G20210A rs1799963, F13A1 G204T rs5985, PROC T8853G, TFPI C536T, MTHFR C677T rs1801133, TNF G(-308)A rs1800629, IL10 A(-1082)G rs1800896, IL10 C(-592)A rs1800872, IL6 C(174)G rs1800795 (Supplementary Table 1) at the Finnish Genome Centre, University of Helsinki, Finland using a Sequenom MassARRAY system, based on MALDI-TOF technology and primer extension chemistry. The SNP assays were designed using Sequenoms Assay Design 2.0 software. The results were analysed with SpectroTYPER software, all genotypes were checked visually twice, and QC checked with in-house developed software. The QC checks included e.g. the analysis of the peak signal morphology and intensity, control DNA, duplicate DNA, and water sample integrity. Two percent of samples were also tested for F5 G1691A rs6025 and F2 G20210A rs1799963 in Finnish Red Cross Blood Service Thrombosis department by Factor V Leiden Kit and Factor II (Prothrombin) G20210A Kit (LightCycler Instrument, Roche Diagnostics GmbH, Mannheim, Germany). There were 78 missing recipient DNA samples and, in addition, 11 recipient DNA samples were only available for FVL and Prothrombin Kit genotyping. All results with both methods were concordant. Forty-three percent of recipients and 44% of donors were typed with the Cytokine Genotyping Kit (One Lambda Inc. CA, USA). Two percent of recipients and donors were typed with both Sequenom and Cytokine Genotyping Kit and these results were concordant. The average genotyping success rate was 99.7% (minmax 98.5100%). Statistical analysis Preliminary power analysis was done with the Genetic Power Calculator [15] based on known baseline characteristics and published results by others of the incidence of studied factors. HardyWeinberg equilibrium was tested with Fishers Exact Test. Descriptive statistics were used to describe the baseline characteristics and duration of follow-up. In bivariate analysis, Fishers Exact Test was used to compare the occurrence of acute rejection, thrombosis, infarction or composite endpoint from the above with respect to categorical baseline characteristics. Multiple corrections were done with Bonferroni correction according to 11 DNA polymorphisms studied in nine candidate genes that give statistically significant P-values less than 0.0045. In addition, the distributions of continuous baseline characteristics with respect to studied complications and available laboratory results after transplantation were compared using a Mann Whitney U-test and a KruskallWallis test. Bivariate survival analysis was done with a KaplanMeier analysis and log rank test. To adjust for confounding factors, multivariate logistic regression and Cox proportional hazards models were used to analyse the risk factors associated with acute rejection, thrombosis, infarction, composite endpoint or first year graft survival after transplantation. Statistical analyses were done using SPSS 13.1 (SPSS Finland Oy, Espoo, Finland) or StatsDirect (StatsDirect Ltd. Cheshire, UK) software. Risk of thrombosis Demographic data of the recipients and donors are given in Table 1. Thrity-eight recipients were diagnosed with venous thrombosis and/or pulmonary embolism (n 14, 37%). Patients aged over the age of 50 years were more likely to suffer a thrombotic complication (OR 2.8, 95% CI 1.35.8). The potential genetic risk factors of the recipients or donors were not significant after multiple corrections (Table 2, data shown for recipient FVL, F2 and MTHFR gene polymorphisms). Warfarin treatment was used by 20 recipients. Warfarin treatment was initiated due to previous thrombosis/embolism (11), atrial fibrillation (5), myocardial infarction (1), aortic valve stenosis (1) or valve Genotype recipient Rejection n (%) No rejection n (%) F5 1691 AA/AG F5 1691 GG F2 20210 AA/AG F2 20210 GG MTHFR 677 CC MTHFR 677 CT/TT Genotype recipient F5 1691 AA/AG F5 1691 GG F2 20210 AA/AG F2 20210 GG MTHFR 677 CC MTHFR 677 CT/TT Genotype recipient F5 1691 AA/AG F5 1691 GG F2 20210 AA/AG F2 20210 GG MTHFR 677 CC MTHFR 677 CT/TT Genotype recipient F5 1691 AA/AG F5 1691 GG F2 20210 AA/AG F2 20210 GG MTHFR 677 CC MTHFR 677 CT/TT Genotype recipient F5 1691 AA/AG F5 1691 GG F2 20210 AA/AG F2 20210 GG MTHFR 677 CC MTHFR 677 CT/TT Some genotype results are missing due to no sample or failed typing. prosthesis (2). None of them had any thromboembolic complications or infarctions. Warfarin treatment was significantly associated with the FVL mutation (P < 0.001, data not shown) but not with other studied polymorphisms. However, all the analyses were done both with and without the warfarin-treated recipients but only results including all the recipients are shown, as omitting patients on warfarin had no statistically significant effect on the results. Risk of infarction There were 48 recipients with infarctions of which 17 (35.4%) were cerebral infarcts, 30 (62.5%) myocardial infarcts and 1 (2.1%) kidney infarct. In addition, two recipients suffered from cerebral infarction after myocardial infarction but only myocardial infarction as the first endpoint was used in analyses. The bivariate analyses of the genetic factors of the recipients and donors influencing infarction did not reveal any significant association after the multiple corrections of the P-values (Table 2, data shown for recipient FVL, F2 and MTHFR gene polymorphisms). Risk of acute rejection episodes Bivariate analyses of the FVL, F2 and MTHFR gene polymorphisms of recipients are shown in Table 2. In the F5 G1961A group (n 20) an acute rejection was determined in six (30%). However, logistic regression analysis proved that none of the genetic factors studied significantly predisposed for acute rejection. Further, in a Cox proportional hazard model no significant genetic factors were found but the same known factors: HLA mismatch [hazard ratio (HR) 2.3, 95% CI 1.63.3], DGF (HR 1.8, 95% CI 1.22.6) and recipients age (HR 0.97, 95% CI 0.96 0.99), which were also found significant in bivariate and in a logistic regression analysis, were significant in the Cox model. Graft survival 1 year after transplantation Graft loss within the first year after transplantation was due to AR, thrombosis or infarct in 28 (62%) out of 45 losses (other graft losses were due to nine patient deaths and eight other reasons e.g. primary disease relapse). The 1-year graft survival was not significantly influenced by the studied genetic factors of recipients or donors analysed by the bivariate (Table 2, data shown for recipient FVL, F2 and MTHFR gene polymorphisms) or survival analysis. None of the clinical factors was significant in bivariate, logistic regression and Cox model analyses. Risk of composite endpoint Composite endpoint included acute rejection, thromboembolic and infarct episodes. If the recipient had more than one complication, only the first was used in composite endpoint analysis. None of the potential genetic risk factors of recipients or donors reached significant P-values after correcting for multiple comparisons (Table 2, data shown for recipient FVL, F2 and MTHFR gene polymorphisms). Furthermore, while KaplanMeier analysis revealed that all the recipients (two patients, one with two transplantations in the study period) with the F5 1691 AA genotype had a complication, the log rank test was not significant when A allele carriers were in the same group (P 0.12). Both TFPI 536 CT carriers also had a complication (P < 0.001). The first patient had a nonfunctioning graft due to a venous occlusion and the second patient had an acute rejection and lower extremity thrombosis. The study aimed to find associations between studied genetic factors and thrombophilia that would help to predict the future risk of acute rejection, thromboembolic, infarct episodes and 1-year graft survival. This information would help in the clinical decision whether to give asymptomatic patients anti-thrombotic prophylaxis or to tailor other medication given to the patient. However, we did not find any associations between the studied genetic mutations and clinical outcomes. We were unable to confirm the previously reported associations between venous thrombosis, rejection, graft survival and FVL, F2 G20210A or MTHFR C677T [2,3] in kidney transplantation recipients. However, our data are similar to those studies which did not find any association in recipients or donors between rejection, thrombosis, infarction or graft survival and FVL, F2 G20210A or MTHFR C677T [2,3,16]. Recently, one large multi-centre study compared the association between the graft survival of cadaver kidney first or retransplants and FVL, F2 G20210A or MTHFR C677T polymorphisms. This study found a suggestive association between first transplants 3-year survival and F2 G20210A polymorphisms [4]. We analysed our data separately for the first and retransplant subgroups and/or immunized or not immunized recipients (PRA <10%) but were unable to find any association with graft survival (data not shown). In addition, a recent large multi-centre study reported that retransplanted, HLA-DRB1 mismatched TNF-308AA carriers had decreased graft survival [17]. As the TNF gene polymorphism is considered to enhance immune responses, it would have been interesting to determine whether this association could be detected in our study material. However, we were unable to perform such an analysis since in our material there were no TNF-308AA carriers among retransplant recipients. Such an analysis would therefore require a larger study population than had been available. We were unable to find associations between the MTHFR C677T polymorphism and studied clinical outcomes. The general practice in our country is to recommend multivitamin use before and after kidney transplantation. This may have masked the possible effects of the MTHFR polymorphism due to sufficient folic acid replacement. The results of previous investigations have been contradictory for several reasons. These include relatively small patient series, heterogeneous genetic background of the study populations and, especially in multi-centre trials, varying treatment protocols. As the Finnish population is genetically highly homogenous, this should allow easier identification of weaker associations between genetic variables and clinical outcome [18]. Thus, one might postulate that if statistically significant genetic associations are not found in a Finnish population, it is less likely that such associations are found in a genetically more mixed Caucasian population. Similarly, it might be inferred that while an association is significant in the Finnish population, this may not be true for a more mixed population. Our investigation was retrospective and focused on determining the relevance of specified candidate gene polymorphisms on clinical outcome. In this respect our study is hampered by the same disadvantages as retrospective studies in general. However, all but one [2] previous investigations have also been retrospective in nature. Further, as our chosen strategy did not include genome-wide analysis of gene polymorphisms, other genes affecting the thrombotic cascade might prove clinically more relevant. Importantly however, with the 11 DNA polymorphisms studied in 9 candidate genes, including the 8 which were not previously studied in kidney transplantation recipients or donors, our investigation is the most comprehensive study of thrombosis-associated polymorphisms in adult kidney transplantation to date. Even though the present study material was relatively large with almost 800 transplantations, it is still limited. We conclude that in a retrospective investigation in adult kidney transplantation setting the studied genetic markers were unable to predict the future risk of acute rejection, thromboembolism, infarct episodes or 1-year graft survival. The routine typing of the studied gene polymorphisms before transplantation does not appear to provide clinical benefit. Supplementary material For Supplementary Online. Material, please refer to NDT Conflict of interest statement. None declared.


This is a preview of a remote PDF: https://ndt.oxfordjournals.org/content/23/1/364.full.pdf

Noora S. Alakulppi, Lauri E. Kyllönen, Jukka Partanen, Kaija T. Salmela, Jarmo T. Laine. Lack of association between thrombosis-associated and cytokine candidate gene polymorphisms and acute rejection or vascular complications after kidney transplantation, Nephrology Dialysis Transplantation, 2008, 364-368, DOI: 10.1093/ndt/gfm528