Portal de Periódicos da CAPES

Assessing the Risk of De Novo Malignancy in Kidney Transplant Recipients

2014; Wolters Kluwer; Volume: 98; Issue: 5 Linguagem: Inglês

10.1097/tp.0000000000000297

1534-6080

Mario Fernández‐Ruiz, Francisco López‐Medrano, Luís M. Allende, Amado Andrés, Estela Paz‐Artal, José María Aguado,

Renal Transplantation Outcomes and Treatments

We have read with interest the recent article by Lim et al., assessing the impact of the use of T-cell–depleting antibodies as antirejection therapy on the development of de novo malignancy after kidney transplantation (KT) (1). They found that recipients who experienced acute rejection (AR) and were treated with T-cell–depleting antibodies exhibited a 1.4-fold increase in the adjusted risk of incident cancer as compared to those who did not experience AR. Interestingly, such an effect was not observed in patients with rejection episodes that did not require therapy with T-cell–depleting agents (1). Following the suggestion made by Snanoudj and Legendre in their editorial (2), we attempted to analyze whether the use of such agents as induction therapy also increases the risk of posttransplant malignancy. In addition, in view that the decrease in antitumor and antiviral T-cell–mediated immune surveillance seems to be the most relevant mechanism underlying the association reported by Lim et al. (1), we explored the potential utility of a monitoring strategy based on some easily available proxy for the posttransplant status of the cell-mediated immunity in an attempt to individualize cancer prevention in this population. All patients undergoing KT at our institution from November 2008 are being enrolled in an ongoing immune-monitoring study in which data on major posttransplant complications are prospectively collected (3, 4). Peripheral blood lymphocyte subpopulations (PBLSs) counts are enumerated at baseline and posttransplant months 1 and 6 using a six-color flow cytometric assay (4). In this substudy, we included patients with a minimum posttransplant follow-up of 90 days, unless death or graft loss occurred earlier. The study outcome was the occurrence of de novo cancer (excluding nonmelanocytic skin cancer) after transplantation. We preestablished two subgroups: those patients who received antithymocyte globulin (ATG) as induction therapy (ATG group), and those who received anti-CD25 monoclonal antibodies or no induction therapy (non-ATG group). Areas under receiver operating characteristic curves (auROC) were used to assess the accuracy of each PBLS count in predicting the development of de novo malignancy. The best cutoff values were then assessed through the calculations of sensitivity and specificity. Survival curves were plotted by the Kaplan-Meier method, and univariate and multivariate Cox regression models were used to evaluate the association between the most predictive parameter identified by auROC analysis and the study outcome. We analyzed 297 patients (162 and 135 in the non-ATG and ATG groups, respectively) throughout a median follow-up of 72.7 months (interquartile rank, 32.8–53.5 months; risk exposure period, 360,248 transplant days). At the end of this period, 255 of them (85.9%) were still alive and with a functioning graft. Seventeen patients (5.7%) were diagnosed with a de novo malignancy within a median of 24.1 months (interquartile rank, 9.2–36.4 months) after transplantation. These malignancies comprised 12 solid tumors, three Kaposi’s sarcomas, two myeloproliferative syndromes, and one posttransplant lymphoproliferative disorder. There were no differences in the malignancy-free survival between ATG and non-ATG groups (Fig. 1A). Of note, only 4 of 73 (5.5%) episodes of AR (occurring in 66 patients) were treated with ATG. The CD8+ T-cell count at month 1 was found to be the most predictive parameter for the subsequent occurrence of posttransplant cancer (auROC, 0.648; P=0.047). By using a cutoff of 0.050×103 cells/μL, we obtained sensitivity, specificity, and positive and negative predictive values of 50.0%, 83.6%, 16.7%, and 96.2%, respectively. One-year and three-year malignancy-free survival rates were significantly lower in patients with a CD8+ T-cell count below this threshold at month 1 (91% and 80% vs. 98% and 94%, respectively; P=0.001) (Fig. 1B). In the Cox model, after adjusting for recipient and donor age, dialysis vintage, number of pretransplant comorbidities, and occurrence of AR within the first 6 months, a CD8+ T-cell count less than 0.050×103 cells/μL at month 1 acted as an independent risk factor for de novo malignancy (hazard ratio, 5.03; 95% confidence interval: 1.82–13.91; P=0.002).FIGURE 1: Kaplan-Meier curves for de novo malignancy-free survival according to: (A) the use of ATG as induction therapy (log-rank P=0.614); and (B) the CD8+ T-cell count at month 1 (log-rank P=0.001). ATG, antithymocyte globulin.Despite the use of ATG as induction therapy in 45.5% of the recipients included in our cohort, we have not been able to demonstrate any discernible direct effect of such a T cell-depleting agent on the incidence of cancer. Although limited by the relatively low number of events, this result suggests that the inflammatory status underlying AR episodes should play a relevant role in the association found in the Australia and New Zealand Dialysis and Transplant Registry (1). On the other hand, the long-lasting dose-dependant T-cell depletion induced by polyclonal antithymocyte preparations (5, 6) offers the opportunity of using the kinetics of certain PBLSs as predictors for de novo cancer development (7). Previous studies have shown a correlation between low CD4+ and CD28+ CD8+ T-cell counts and the risk of posttransplant malignancy in kidney (8, 9) and liver transplant recipients (10), respectively. We propose a cutoff for CD8+ T-cell count at month 1 with an excellent negative predictive value for predicting the subsequent occurrence of cancer. Although the positive predictive value was suboptimal, monitoring of this PBLS may effectively discriminate low-risk patients, thus allowing to focus screening and prevention efforts on those KT recipients with CD8+ T-cell counts below the protective threshold. Further studies with larger patient samples might help to identify whether monitoring of specific PBLSs according to the type of induction therapy used could increase the predictive accuracy of this approach. Mario Fernández-Ruiz 1 Francisco López-Medrano1 Luis M. Allende2 Amado Andrés3 Estela Paz-Artal2 José María Aguado1 1 Unit of Infectious Diseases Instituto de Investigación, Hospital Universitario 12 de Octubre Madrid, Spain 2 Department of Immunology Instituto de Investigación, Hospital Universitario 12 de Octubre Madrid, Spain 3 Department of Nephrology Instituto de Investigación, Hospital Universitario 12 de Octubre Madrid, Spain