Objectives: Allograft rejection is an important cause of early and long-term graft loss in kidney transplant recipients. Tumor necrosis factor-alpha promotes T-cell activation, the key reaction leading to allograft rejection. Here, we investigated whether serum and urinary tumor necrosis factor-alpha levels can predict allograft rejection.
Materials and Methods: This study included 65 living related-donor renal transplant recipients with mean follow-up of 26 ± 9 months. Serum and urinary tumor necrosis factor-alpha levels were measured at pretransplant and at posttransplant time points (days 1 and 7 and months 3 and 6); serum creatinine levels were also monitored during posttransplant follow-up. Standard enzyme-linked immunoabsorbent assay was used to detect tumor necrosis factor-alpha levels. Clinical variables were monitored.
Results: Nine of 65 patients (13.8%) had biopsy-proven rejection during follow-up. Preoperative serum and urinary tumor necrosis factor-alpha levels were not significantly different when we compared patients with and without rejection. Serum tumor necrosis factor-alpha levels (in pg/mL) were significantly higher in the allograft rejection versus nonrejection group at day 7 (11.5 ± 4.7 vs 15.4 ± 5.8; P = .029) and month 1 (11.1 ± 4.8 vs 17.8 ± 10.9; P =.003). Urinary tumor necrosis factor-alpha levels (in pg/mL) were also elevated in the allograft rejection versus the nonrejection group at days 1 (10.2 ± 2.5 vs 14.1 ± 6.8; P = .002) and 7 (9.8 ± 2.2 vs 14.5 ± 2.7; P < .001) and at months 1 (8.0 ± 1.7 vs 11.8 ± 2.4; P < .001), 3 (7.7 ± 1.6 vs 9.6 ± 1.7; P = .002), and 6 (7.4 ± 1.6 vs 8.9 ± 0.9; P = .005).
Conclusions: Our preliminary findings suggest that tumor necrosis factor-alpha has a role in diagnosing renal transplant rejection. Serum and urinary tumor necrosis factor-alpha levels may be a possible predictor for allograft rejection.
Key words : Cytokine, Graft rejection, Kidney transplantation
Renal transplant is the best treatment for end-stage renal disease. Acute rejection is one of the most important causes of allograft dysfunction and can lead to early and long-term graft loss in kidney transplant recipients despite antirejection therapy.1 Younger recipients, older donors, donor-recipient human leukocyte antigen (HLA) mismatches, pretransplant anti-HLA antibodies, panel reactive antibodies, and the adequacy of baseline immuno-suppression are major predictors of acute rejection.2 Transplant renal biopsy is the criterion standard for diagnosing acute rejection and guiding corrective therapy; however, many centers do not routinely consider graft biopsy at the onset of renal dysfunction.3 A noninvasive test would be beneficial for diagnosing acute renal rejection.
Tumor necrosis factor-alpha (TNF-α) is a proinflammatory cytokine produced by monocytes/-macrophages that binds to TNF receptors on endothelial or tubular cells. Tumor necrosis factor-alpha activates lymphocytes and antigen-presenting cells and increases expression of the MHC class II genes and intercellular adhesion molecules on the cell surface, which causes cells to undergo apoptosis.4 The role of TNF-α in allograft rejection has been previously well defined, with animal studies also showing the important role of TNF-α in ischemia-reperfusion injury progression.5 However, the prognostic value of TNF-α has so far not been defined. The aim of this study was to determine the diagnostic value of serum and urine TNF-α levels in allograft rejection.
Materials and Methods
Study population and procedures
This was a prospective cohort study that analyzed patients from a renal transplant unit. The study protocol was approved by the Istanbul University Research Ethics Committee of the Medical Faculty, and written informed consent was obtained from all participants.
Sixty-five patients (61.5% male; mean age of 36.78 ± 12.94 y) who underwent living-donor renal transplant between 2013 and 2015 were included. Patients who had follow-up of more than 3 months after renal transplant were included in this analysis. All transplant procedures were conducted after a negative complement-dependent cytotoxicity cross-match. Patients who had a positive complement-dependent cytotoxicity crossmatch or flow cytometry test were excluded. Data of transplant recipients were recorded prospectively in a large database.
The primary outcome variable was biopsy-proven rejection during follow-up in renal transplant patients.
Blood and urinary samples were collected before and after transplant at days 1 and 7; at months 1, 3, and 6; and at time of rejection. The blood samples were centrifuged at 2600g for 10 minutes. Serum and urinary samples were stored until analysis day at -80°C. Serum and urinary TNF-α levels were measured at posttransplant days 1 and 7; at months 1, 3, and 6; and at the time of rejection. Serum creatinine was also monitored during post-transplant follow-up. Standard enzyme-linked immuno-absorbent assay (Invitrogen Diagnostics, Life Technologies, Carlsbad, CA, USA) was used for detection of TNF-α levels.
Clinical variables that were monitored included rejection episodes and stable graft function. A rejection episode was defined based on clinical or biopsy findings according to the Banff criteria.6 Clinical rejection was identified by increased creatinine levels in the absence of infection, obstruction, or evidence of drug toxicity. Acute rejection episodes were treated with a high daily dose of intravenous methylprednisolone (500 mg each dose) for 3 days; in refractory cases, patients were given antithymocyte globulin (2.5-3 mg/kg/day) for 10 to 12 days.
Data analysis was performed using the statistical software package SPSS (SPSS: An IBM Company, version 21, IBM Corporation, Armonk, NY, USA). Data are shown as means and SD. Statistical analysis was performed using Mann-Whitney U test. For categorical data, the Fisher exact test was used. A multivariate regression analysis was performed for demographic variables. The receiver operating characteristic curve analysis was used to determine the diagnostic power (area under the curve, sensitivity, and specificity) of TNF-α for rejection. P values < .05 were considered statistically significant.
Sixty-five patients who underwent living related-donor renal transplant were included in the study. Mean (SD) ages of patients in the rejection and nonrejection groups were 34.67 (6.83) and 37.13 (13.68) years, respectively. Demographic and clinical characteristics of renal transplant patients are summarized in Table 1.
All patients were classified regarding graft function. Serum creatinine levels were higher in the rejection group than in the nonrejection group at day 7 (P = .001) and at month 1 (P = .004). No significant differences between other periods were observed (Figure 1).
All parameters were analyzed in relation to allograft outcome. Also, we investigated associations between age, warm and cold ischemia time, and number of HLA mismatches and allograft outcomes for each group (P > .05). Renal transplant patients were on a standard triple drug immunosuppressive regimen, consisting of calcineurin inhibitors (tacrolimus) with either mycophenolate mofetil or mycophenolate sodium and steroid (prednisone). These patient characteristics are shown in Table 2.
We examined 390 urinary and serum samples from patients, which included samples from 9 patients with rejection and samples from 56 patients without rejection. All parameters were analyzed in relation to allograft outcome, and patients were classified according to their transplant outcomes in terms of rejection. Nine patients (13.8%) had biopsy-proven acute rejection during follow-up (with 56 patients [86.2%] included in the nonrejection group). Urinary and serum TNF-α levels were assessed for each group.
Regarding preoperative levels of serum and urinary TNF-α, we observed no significant differences between patients with and without rejection. Urinary levels of TNF-α increased to 18.2 pg/mL in patients with rejection at time of biopsy-proven rejection (P < .001), and serum levels of TNF-α were increased to 22.1 pg/mL in patients with rejection at time of biopsy-proven rejection (P < .001).
Serum TNF-α levels were significantly higher in the allograft rejection group at day 7 (P = .006) and month 1 (P = .042); urinary TNF-α levels were also elevated in the allograft rejection group versus the nonrejection group at days 1 (P < .001) and 7 (P < .001) and at months 1 (P < .001), 3 (P = .011), and 6 (P = .037). The urinary TNF-α levels for both groups are summarized in Table 3.
Regarding predictive factors, the multivariate logistic regression analysis revealed that the following outcomes were not predictive of rejection: age, sex, weight, anti-HLA antibody, immunosup-pressive regimen, cold ischemia and warm ischemia time, and dialysis duration (measured at days 1 and 7 and at months 1, 3, and 6 after transplant). Results of the multivariate analysis for rejection in the study groups are shown in Table 4.
The receiver operating characteristic analyses showed that TNF-α was a modestly better predictor. The highest sensitivity (100%) and specificity (84%) were observed in the rejection group versus no rejection group at a cutoff value of 11.03 ng/mL of urinary TNF-α level, and the area under the curve was 0.947 (95% confidence interval, 0.877-1.000) (Table 5).
Acute rejection after renal transplant remains a major clinical problem that affects both patient and graft survival. Most cases of rejection occur within a few weeks after transplant.7 Sensitive and rapid assessments of changes in graft function are of primary importance in renal transplant recipients in the early posttransplant period. Although early rejection episodes, predicted by urinary markers, may be prevented with therapeutic interventions, it remains a major risk factor for subsequent graft dysfunction.8,9
Among the patients who had renal allograft rejection, 83% occurred within 3 months after transplant and 66.7% occurred within 1 month after transplant. Many researchers have reported increased serum concentrations of TNF-α during acute rejection of liver, heart, and kidney allografts.10,11
This study demonstrated that significant changes in TNF-α levels in urine and serum, in the early renal posttransplant period, can be detected and that some of these changes correlate with allograft rejection. In 9 of 65 patients with biopsy-proven rejection during our 26-month follow-up, TNF-α levels in urinary samples were significantly increased at days 1 and 7 and at months 1, 3, and 6. Serum TNF-α levels were also significantly increased in patients with rejection at day 7 and at month 1.
Tumor necrosis factor-alpha also seems to play a central role in the immune response to alloantigen, with several studies demonstrating increased TNF-α levels in patients with liver, kidney, and pancreas transplant rejection.12 Although few studies have associated serum levels of TNF-α with acute rejection after renal transplant, Ficek and associates reported markedly increased TNF-α levels in patients with acute renal failure.13 Budak and associates also reported significantly higher serum TNF-α receptor levels on day 1 and month 1 in patients with acute rejection.14
We found that clinical variables such as sex, age, HLA compatibility, anti-HLA antibody, and immunosuppressive regimen had no significant effects on rejection. However, Ding and associates, in their investigation on whether clinical variables significantly effected acute rejection, found that panel reactive antibody levels > 10% were significantly associated with acute rejection.15
In our study, area under the curve values showed that urinary TNF-α levels were better than serum TNF-α levels and serum creatinine for the determination of rejection. Budak and associates14 used receiver operating characteristic analysis to estimate acute rejection and found that serum TNF-α receptor levels determined on day 7 and month 1 have the highest sensitivity and specificity. Therefore, urine TNF-α levels may be a better predictor of rejection together with serum creatinine. In addition, the observed increased levels of TNF-α support its role as a mediator of immune response during rejection.
Our present study shows that long-term stable renal transplant patients have many suppressed immune parameters, such as high TNF-α urinary levels, which may be beneficial for immunosuppression and maintenance of allograft acceptance. We conclude that measuring TNF-α in urine may be a surrogate marker of immunosuppression in renal transplant patients.
Volume : 16
Issue : 6
Pages : 671 - 675
DOI : 10.6002/ect.2017.0166
From the 1Department of Medical Biology, the 2Division of Nephrology, Department
of Internal Medicine, the 3Department of Anesthesia, and the 4Department of
Urology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey
Acknowledgements: This study was presented as poster presentation at the 31st European Immunogenetic and Histocompatibility Conference. The study was supported by the Scientific Research Projects Coordination Unit (BAP) of Istanbul University with project number 47549. The authors have no conflicts of interest to declare.
Corresponding author: Hayriye Senturk Ciftci, Department of Medical Biology, Istanbul Medical Faculty, Istanbul University, Capa, Fatih, Istanbul, Turkey
Phone: +90 532 316 4576
Table 1. Demographic Details of Renal Transplant Patients
Table 2. Characteristics of Patients in the Rejection and Nonrejection Groups
Table 3. Tumor Necrosis Factor-Alpha Levels in Study Groups
Table 4. Multivariate Logistic Regression Analysis of Potential Risk Factors for Rejection
Table 5. Serum Creatinine and Serum and Urine Tumor Necrosis Factor-Alpha Results for Determination of Rejection After Transplant
Figure 1. Comparison of Serum Creatinine Levels Before and After Transplant