Cytomegalovirus (CMV) infection still remains as the most common opportunistic infection after kidney transplantation.1 Both indirect and direct effects associated with CMV infection negatively challenge the course of the transplant organ, favoring allograft rejection, as well as directly threatening patient survival.2,3
Although the introduction of novel preventive strategies, both prophylaxis and preemptive have significantly shown its efficacy preventing CMV infection as well as improving transplant outcomes,4,5 the decision of the type and duration of these therapies is not well defined. Currently, high serologically risk patients (D+/R−) as well as those transplant recipients receiving T cell induction therapy are considered to be at particularly high risk (HR) of CMV infection thus, either a 6- or 3-month course of antiviral prophylaxis therapy, respectively, is highly recommended. However, since an important proportion of seropositive (R+) patients not receiving T cell–depleting antibodies will also develop CMV infection (25-40%), a short course of antiviral prophylaxis therapy is also performed in many transplant programs.4 Furthermore, despite the reduced incidence of CMV infection using antiviral prophylaxis, late-onset CMV infection after prophylaxis withdrawal remains as main posttransplant complication, favoring higher rates of mortality and graft loss.6,7 Therefore, accurate biomarkers allowing a precise identification of transplant patients requiring antiviral prophylaxis therapy as well as its duration to avoid late-onset CMV infection is highly warranted.
Cell-mediated immunity (CMI) against CMV, both CD4+ and CD8+ T cells, has a critical role controlling and restricting viral replication.8,9 Interesting recent reports have shown the value of evaluating the CMV-specific CMI using different immune assays such as the Quantiferon ELISA-based assay, the IFN-γ enzyme-linked ImmunoSpot assay (ELISpot) and intracellular-cytokine flow cytometry10-15 identify at-risk patients of CMV infection at different time-points of transplantation.16-18 However, whether the type of induction immunosuppressive therapy used, either T cell–depleting antibodies or monoclonal antibodies, may be a relevant factor inducing different risk of late-onset CMV infection after prophylaxis cessation has not been thoroughly investigated.
Herein, we prospectively analyzed in 2 groups of 96 consecutive kidney transplant patients receiving either rabbit antithymocyte globulins (rATG) or anti-IL2 receptor antibody (IL2RA) induction therapy, the usefulness of monitoring CMV-specific CMI using a sensitive IFN-y ELISpot assay against 2 main dominant immunogenic CMV antigens, pp65 (phosphoprotein 65) and IE-1 (immediate-early 1) at the time of 3-month antiviral prophylaxis withdrawal, to identify at-risk patients of late-onset CMV. Moreover, the analysis of the kinetics of CMV-specific CMI both at baseline and at different time-points prior to prophylaxis cessation has been investigated to assess the proportion of patients achieving a protective CMI threshold during the course of prophylaxis therapy according to the type of induction therapy used.
MATERIALS AND METHODS
Patients and Study Design
From December 2014 until November 2016, a 12-month prospective, observational study was conducted in 96 consecutive adult ABO-compatible, CMV(IgG)-seropositive kidney transplant recipients (R+), receiving a CMV(IgG)-seropositive kidney allograft (D+) to assess the impact of CMV-specific CMI at the time of prophylaxis cessation on late-onset CMV infection. Two groups of 96 consecutive patients were evaluated according to the type of induction therapy received: either anti-IL2RA (n = 50) or rATG (n = 46). Patients with high immunological risk factors (retransplant patients, patients with preformed anti-HLA antibodies) and/or received kidney allograft from donors after cardiac death received rATG as induction therapy whereas all low-risk (LR) transplant patients received anti-IL2RA induction therapy. All patients received a 3-month course of antiviral prophylaxis therapy with valgancyclovir, followed by secondary preemptive therapy with systematic screening of CMV viremia every 2 weeks until month 6 after transplantation (Figure 1). Among the 96 patients enrolled in the study, none of them prematurely stopped the antiviral therapy. However, in 7 patients within the rATG-treated group, Valgancyclovir doses had to be reduced because of leukopenia and/or low platelet counts. Likewise, Valgancyclovir doses were reduced in 3 patients within the anti-IL2RA-treated group. Importantly, only 1 out of 7 rATG-treated patients and none of the 3 patients in the anti-IL2RA group developed late-onset CMV infection. Additionally, patients in whom blood samples could be obtained were also monitored for their CMV-specific CMI at different time-points prior to prophylaxis withdrawal: at baseline (n = 87), at 2 weeks (n = 79) and at 1 month (n = 83) after transplantation. All patients included in the study gave written informed consent to participate and the study was approved by the ethics committee at Bellvitge University Hospital.
Main outcomes of the study were the advent of CMV infection and disease after prophylaxis withdrawal at 3 months after transplantation. Both clinical events were defined according to the recently reported guidelines by the CMV drug Development Forum.19 CMV infection was considered when any CMV viral nucleic acid was detected in plasma. CMV disease was defined as evidence of CMV replication with compatible symptoms, including both viral syndrome and invasive tissue disease.
Serology and Microbiological Tests
Pretransplant CMV serological status of donors and recipients was determined using a commercial Human Anti-CMV IgG Enzyme-Linked ImmunoSorbent Assay (ELISA) kit (BioCheck, Maine) according to manufacturer's instructions. Surveillance of quantitative CMV DNA detection was done using a real-time CMV kit (Abbott, Abbott Park, IL, USA). The lower limit of detection for this PCR kit was 100 UI/mL.
Assessment of CMV-specific CMI
CMV-specific CMI was evaluated with an IFN-γ enzyme-linked immunosorbent spot assay (T-spot.CMV, Oxford Immunotec, Inc) against 2 major CMV antigens, the immediate-early protein 1 (IE-1) and 65 kDa phosphoprotein (pp65) using overlapping peptide pools covering the whole antigen length. Peripheral blood mononuclear cells (PBMC) were isolated and cryopreserved in liquid nitrogen for its subsequent use to perform the T-spot. CMV ELISpot assay. Briefly, 3 × 105 PBMCs (in 100 μL) were stimulated with CMV antigens for 18-20 hours in duplicate wells. IFN-γ spots were detected after using biotinylated antihuman IFN-γ antibody plus the addition of alkaline phosphatase conjugate substrate (Oxford Immunotec, lnc). Spots obtained were counted with an ELISpot reader (AID GmbH, ELISpot Reader HR fourth generation). In each test, AIM-V medium alone and Phytohemagluttinin (Oxford Immunotec, Inc, Oxford, UK) were used as negative and positive controls, respectively.
All data are presented as mean ± standard deviation. Groups were compared using the χ2 test for categorical variables, the 1-way analysis of variance or Student t test and paired samples t test for normally distributed data, and the nonparametric Kruskal-Wallis or Mann-Whitney U test for non-normally distributed variables. CMV infection and disease were considered outcome variables of the study. Univariate Cox regression analyses were performed to evaluate the association between CMV-specific CMI and CMV infection. Results were expressed as Hazard ratios (H-R) with 95% confidence intervals (CI). Binary logistic regression analyses were performed to determine the independent correlation of variables that were statistically significant in the univariate analyses and those potentially associated with the presence of posttransplant CMV infection. The statistical significance level was defined as 2-tailed P value less than 0.05. All statistical analyses were performed with IBM Spss Statistics (version 23) and GraphPad Prism (version 6.0; GraphPad Software, San Diego, CA).
Demographic and Clinical Outcome of the Study Population
Main demographic and baseline clinical characteristics of all patients of the study are depicted in Table 1. Only (IgG) CMV-seropositives (R+) receiving a seropositive allograft (D+) were considered in this study. Most patients received a deceased kidney donor (87.5%) and were all treated with the calcineurin inhibitor (CNI) tacrolimus (TAC), combined with mycophenolate mofetil (MMF). The incidence of biopsy-proven acute rejection (BPAR) was 8.3% (8/96). One (1%) of 96 patients of the study died with a functioning graft. Fourteen (14.6%) of 96 and only 2 (2.1%) of 96 patients developed late-onset CMV infection and disease (both with gastrointestinal involvement), respectively. Mean time between CMV infection and prophylaxis cessation was 54.14 ± 41.48 days among anti-IL2RA–treated patients and 47.86 ± 23.07 days in those receiving rATG (P = 0.732). Mean peak of CMV PCR load in patients developing late-onset CMV infection was 4865.86 ± 7319.95 IU/mL.
Association Between Late-onset CMV Infection and CMV-specific CMI at 3-month Prophylaxis Withdrawal
At 3 months after transplantation, antiviral prophylaxis therapy was stopped in all patients and subsequent systematic surveillance CMV DNAemia was assessed. As shown in Figures 2A-B, transplant recipients developing late-onset CMV infection displayed significantly lower IFN-γ–producing T-cell frequencies against both IE-1 and pp65 CMV antigens than patients that did not. Likewise, the 2 patients developing late-onset CMV disease did also show significantly lower CMV (IE-1)-specific CMI than patients that did not.
Mean CMV-specific CMI against either IE-1 or pp65 was not different between rATG and anti-IL2RA-treated patients (152.78 ± 189.10 vs 179.21 ± 234.99 IFN-γ spots, respectively, for IE-1, P = 0.547 and 304.61 ± 234.62 vs 313.18 ± 273.44 IFN-γ spots, respectively, for pp65, P = 0.870). When patients were stratified according to the type of induction therapy, patients developing late-onset CMV infection showed significantly lower CMV-specific CMI, particularly against IE-1 CMV antigens, regardless the type of induction therapy (Figures 2C-D).
Stratification of CMV-specific CMI for CMV Infection Risk Evaluation
To obtain an easy CMV-specific CMI risk stratification score, we performed sensitivity/specificity receiver operating characteristic (ROC) curve analyses to obtain a specific CMI cutoff value discriminating transplant patients at different risks of late-onset CMV infection. As shown in Figures 3A, two specific cutoffs, 25 and 130 IFN-γ spots/3 × 105 PBMC for IE-1 and pp65, respectively, were the most accurate values discriminating CMV infection (areas under the curve [AUC], 0.704; 95% CI, 0.585-0.823; P = 0.015 for IE-1 and AUC, 0.680; 95% CI, 0.540-0.820; P = 0.032 for pp65) with 71.4% and 57.1% sensitivity, 70.7% and 75.6% specificity, 93.5% and 91.2% negative predictive value (NPV) and 29.4% and 28.6% positive predictive value (PPV) for IE-1 and pp65, respectively (Table 2). The same cutoffs and predictive values were observed when analyzing patients according to the different induction therapies used (Figure S1, SDC, http://links.lww.com/TP/B629). Using these cutoffs, patients were then classified into 3 different risk groups: HR if a negative test against both antigens; LR if positive tests against both antigens, and intermediate risk (IR) if a negative result against 1 of the 2 CMV antigens. As illustrated in Figure 3B, 20 (21%) of 96 patients classified as HR, 54 (56%) of 96 patients as LR and 22 (23%) of 96 patients as IR.
When Kaplan-Meier CMV infection-free survival curves were done using these stratification scores, either individually for each antigen or combining both antigens, a significantly higher probability of late-onset CMV infection was observed in patients with a negative IE-1 CMI (Log-rank test = 0.002) (Figure 4A) or a negative pp65 CMI (Log-rank test = 0.014) (Figure 4B). When taking into account the CMI result against both CMV antigens, the risk of late-onset CMV infection increased gradually according to the 3 different risk score groups (Log-rank test = 0.006) (Figure 4C): 7 (35%) of 20 within HR patients, 4 (18.2%) of 22 within IR and 3 (5.6%) of 54 within LR patients (H-R, 4.084; 95% CI, 1.431-11.651; P = 0.009, for HR; H-R, 1.477; 95% CI, 0.463-4.712, P = 0.510, for IR and H-R, 0.189; 95% CI, 0.053-0.679; P = 0.011, for LR). As depicted in Table 3, the predictive capacity of this qualitative combined CMI risk score maintained a similar NPV and specificity, but decreased the sensitivity of the test as compared with the IE-1-specific CMI. Notably, 3 (75%) of 4 IR patients developing late-onset CMV infection were IE-1-CMI negative and pp65-CMI positive, whereas only 1 patient developing CMV infection showed a negative pp65-CMI and a positive IE-1-CMI.
No statistical significant differences were observed in the peak of CMV PCR load between the different CMI risk groups (data not shown).
CMV-specific CMI Stratification Independently Predicts Late-onset CMV Infection
A multivariate binary logistic regression analysis evaluating main clinical, demographic and immunological variables predicting CMV infection revealed that CMV-specific CMI, defined either as the HR CMI variable or using the binary CMI result against IE-1 CMV antigen at the time of prophylaxis cessation, were the only independent correlates predicting late-onset CMV infection (Table 4).
CMV-specific CMI Assessed at Previous Time Points of Prophylaxis Cessation Discriminates Patients that May Benefit of Earlier Prophylaxis Withdrawal
We next evaluated the kinetics of CMV-specific CMI at 3 different time points prior to prophylaxis cessation: at baseline, at 15 days and 1 month after transplantation. As illustrated in Figure 5, a profound abrogation of CMV-specific CMI, particularly against IE-1 CMV antigens, was predominantly observed among rATG-treated patients during the first 3 months after transplantation as compared to patients receiving anti-IL2RA, although a significant inhibition of CMV-specific CMI did also occur among this latter group as compared to baseline. However, as previously shown, CMV-specific CMI achieved the same levels in both groups of induction treatments at 3 months after transplantation. The transient but profound CMV-specific CMI abrogation among rATG-treated patients is clearly illustrated by the poor correlation between pretransplant CMV-specific CMI, especially against IE-1, and the different posttransplant time points. Conversely, a strong positive correlation was observed among anti-IL2RA-treated patients (Figures S2A and S2B, SDC, http://links.lww.com/TP/B629).
Finally, the different CMI risk scores conferring protection against late-onset CMV infection obtained at 3 months were used at previous time points during the course of antiviral prophylaxis therapy as well as at baseline. As shown in Figure 6, although most anti-IL2RA–treated patients maintained the same CMI risk score at all time points, a number of them (20%) switched to a HR score from baseline until month 1 after transplantation while recovering their original protected CMI status at month 3. Conversely, a high proportion of patients receiving rATG (50%) switched to a HR score after transplantation, and recovered their protected CMI status at month 3 after transplant.
While the use of antiviral prophylaxis after kidney transplantation is particularly recommended among transplant recipients receiving T cell–depleting agents and patients with a high-risk donor/recipient paired serostatus (D+/R−), a broader use of such preventive strategy is carried out in many transplant programs. Nevertheless, the advent of late-onset CMV infection may appear in a significant proportion of patients after prophylaxis withdrawal. Herein, we aimed to prospectively evaluate the value of assessing CMV-specific CMI at the time of prophylaxis withdrawal in 2 consecutive groups of IR serological kidney transplant patients (D+/R+), receiving either anti-IL2RA or rATG. Furthermore, the assessment of specific CMI risk stratification scores for its easy clinical use at the time of prophylaxis cessation as well as at previous time points has been thoroughly investigated.
The use of an IFN-γ ELISpot assay assessing CMV-specific CMI against 2 major immunogenic CMV antigens at the time of 3-month prophylaxis cessation revealed that patients developing late-onset CMV infection displayed significantly lower IFN-γ-producing T cell frequencies, particularly against the IE-1 CMV antigen. Subsequently, a ROC curve analysis identified 2 specific and sensitive CMI cutoffs against each CMV antigen (25 and 130 IFN-γ spots for IE-1 and pp65, respectively) that classified transplant patients at HR or LR of late-onset CMV infection. To test whether the combination of both CMI responses could improve its predictive value, both CMV-antigen CMI responses were merged and the predictive capacity of the CMI risk score combination showed similar NPV to IE-1 and pp65 (90%), and higher specificity (>80%). Indeed, the cumulative incidence of late-onset CMV infection was significantly higher among patients with a negative CMI against either IE-1 or pp65, similarly to those with a HR CMI risk score.
This data was reproducible when stratifying patients according to the type of induction therapy received, suggesting an optimal recovery of CMV-specific CMI at 3 months after transplantation among R+ kidney transplant patients receiving rATG. Even rATG has been previously associated with higher rates of CMV infection,20 differences between rATG-treated patients and anti-IL2R group were not observed. These results could be explained by the fact that most patients receiving rATG showed a protective CMI at baseline, and this protective state might have been recovered due to homeostatic proliferation after the 3-month course of universal prophylaxis.
The multivariate binary logistic regression models using either the combination of CMI risk score or each CMV antigen individually, revealed that either IE-1-specific CMI or a HR CMI risk score were independent correlates of late-onset CMV infection, showing the former the strongest prediction. This data is in line with previous studies from our group reporting the preponderance of this specific CMV-specific CMI discriminating transplant patients at increased risk of CMV infection.13,18 While recent reports using the Quantiferon test12,16 have also showed the value of monitoring CMV-specific CMI at the time of prophylaxis cessation, they have predominantly focused on high serological risk patients (D+/R−) and not discriminating the type of induction therapy used. In our study, no undetermined results were observed when using ELISpot and, these results, in addition to the fact that both CD8+ and CD4+ T cells are assessed using overlapping peptide pools avoiding potential HLA restrictions confer a relevant advantage to the ELISpot test.
The analysis of CMV-specific CMI at previous time-points of prophylaxis cessation as well as at baseline clearly showed the profound abrogation of CMV-specific CMI in patients receiving T-cell depletion therapy. However, it is also important to highlight that patients receiving anti-IL2RA induction do also experience a certain CMV-specific CMI inhibition, although far less pronounced. Thus, suggesting early phases after transplantation as the better time point for monitoring CMV-specific CMI for a more accurate prediction of infection risk in patients receiving anti-IL2RA induction therapy. Conversely, the profound rATG-driven CMI abrogation after transplantation in the great majority of patients, strongly suggests the need of providing an initial course of antiviral prophylaxis therapy until the subsequent recovery of CMV CMI, more likely after month 3 after transplantation.
This study has some limitations. First, the observational nature of the study, without therapeutic intervention, may somehow weaken our data. However, the large and well-defined groups of patients evaluated as well as the consistency of the results obtained significantly counterbalance this constraint. Moreover, the absence of CMI evaluation at month 2 after transplantation could have contributed to a better understanding of the progressive recovery of CMV-specific CMI. Finally, the low number of CMV infections and CMV disease events might be a limitation, however, these data is in line with low incidence of late-onset CMV infection and disease events broadly reported among D+/R+.6,21
In conclusion, we have shown that assessing CMV-specific CMI with a sensitive IFN-γ ELISpot assay at the time of prophylaxis withdrawal is an accurate tool to help transplant clinicians identifying at-risk patients of developing late-onset CMV infection and disease and thus, help establishing guided decision-making regarding whether maintaining a longer course of prophylaxis therapy or conversely, safely stop this antiviral treatment. Furthermore, our results strongly suggest that unlike transplant recipients receiving rATG, anti-IL2RA-treated patients may benefit from an earlier evaluation of their CMV-specific CMI, to help establishing the type of preventive therapy to follow soon after transplantation.
The authors acknowledge our Bio-bank unit at IDIBELL and Ms Gema Cerezo and Ms Pilar Arana in our laboratory for carefully processing and managing all biological samples as well as to our study coordinators Dr Carolina Polo and Ms Eulàlia Molinas for organizing all biological samples collection. The authors also thank CERCA programme/Generalitat de Catalunya for institutional support.
1. Fishman JA. Infection in solid-organ transplant recipients. N Engl J Med
2. Sagedal S, Nordal KP, Hartmann A, et al. The impact of cytomegalovirus infection and disease on rejection episodes in renal allograft recipients. Am J Transplant
3. Sagedal S, Hartmann A, Nordal KP, et al. Impact of early cytomegalovirus infection and disease on long-term recipient and kidney graft survival. Kidney Int
4. Kotton CN, Kumar D, Caliendo AM, et al. The third international consensus guidelines on the management of cytomegalovirus in solid-organ transplantation. Transplantation
5. Witzke O, Nitschke M, Bartels M, et al. Valganciclovir prophylaxis versus preemptive therapy in cytomegalovirus-positive renal allograft recipients: long-term results after 7 years of a randomized clinical trial. Transplantation
6. Limaye AP, Bakthavatsalam R, Kim HW, et al. Late-onset cytomegalovirus disease in liver transplant recipients despite antiviral prophylaxis. Transplantation
7. Humar A, Lebranchu Y, Vincenti F, et al. The efficacy and safety of 200 days valganciclovir cytomegalovirus prophylaxis in high-risk kidney transplant recipients. Am J Transplant
8. Walter EA, Greenberg PD, Gilbert MJ, et al. Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N Engl J Med
9. Sacre K, Carcelain G, Cassoux N, et al. Repertoire, diversity, and differentiation of specific CD8 T cells are associated with immune protection against human cytomegalovirus disease. J Exp Med
10. Lúcia M, Crespo E, Cruzado JM, et al. Human CMV-specific T-cell responses in kidney transplantation; toward changing current risk-stratification paradigm. Transpl Int
11. Sester M, Leboeuf C, Schmidt T, et al. The “ABC” of virus-specific T cell immunity in solid organ transplantation. Am J Transplant
12. Kumar D, Chernenko S, Moussa G, et al. Cell-mediated immunity to predict cytomegalovirus disease in high-risk solid organ transplant recipients. Am J Transplant
13. Lúcia M, Crespo E, Melilli E, et al. Preformed frequencies of cytomegalovirus (CMV)-specific memory T and B cells identify protected CMV-sensitized individuals among seronegative kidney transplant recipients. Clin Infect Dis
14. Bestard O, Nickel P, Cruzado JM, et al. Circulating alloreactive T cells correlate with graft function in longstanding renal transplant recipients. J Am Soc Nephrol
15. Schachtner T, Stein M, Reinke P. CMV-specific T cell monitoring offers superior risk stratification of CMV-seronegative kidney transplant recipients of a CMV-Seropositive donor. Transplantation
16. Manuel O, Husain S, Kumar D, et al. Assessment of cytomegalovirus-specific cell-mediated immunity for the prediction of cytomegalovirus disease in high-risk solid-organ transplant recipients: a multicenter cohort study. Clin Infect Dis
17. Kumar D, Mian M, Singer L, et al. An interventional study using cell-mediated immunity to personalize therapy for cytomegalovirus infection after transplantation. Am J Transplant
18. Bestard O, Lucia M, Crespo E, et al. Pretransplant immediately early-1-specific T cell responses provide protection for CMV infection after kidney transplantation. Am J Transplant
19. Ljungman P, Boeckh M, Hirsch HH, et al. Definitions of cytomegalovirus infection and disease in transplant patients for use in clinical trials: Table 1. Snydman DR, editor. Clin Infect Dis
20. Brennan DC, Daller JA, Lake KD, et al. Rabbit antithymocyte globulin versus basiliximab in renal transplantation. N Engl J Med
21. Jamal AJ, Husain S, Li Y, et al. Risk factors for late-onset cytomegalovirus infection or disease in kidney transplant recipients. Transplantation