Of the two patients with a positive CMI (interferon-γ at baseline 0.39 and 1.14 IU/mL, respectively) who had disease progression, one was a CMV donor seropositive/recipient seropositive (D+/R+) kidney transplant (peak viral load=11,100 copies/mL) and the other was a D+/R+ lung transplant patient (peak viral load=18,850 copies/mL), both of whom developed viral syndrome.
Interferon-γ production was also analyzed as continuous variable. In patients who had spontaneous clearance, the baseline interferon-γ was a median of 1.73 IU/mL (IQR 0.24–10.87) vs. 0.05 IU/mL (IQR 0.01–0.32) for patients with virologic or clinical progression (P=0.003). At the second time point, median interferon-γ levels were 3.75 vs. 0.07 IU/mL in the two groups, respectively; P=0.004. By the third time point the median levels were 3.49 vs. 0.08 IU/mL; P=0.038. A receiver operator characteristics curve analysis was performed for prediction of spontaneous clearance of viremia based on CMV-specific interferon-γ levels. The area under the curve (AUC) was 0.843 (P value for difference from chance alone [AUC=0.5] was 0.003; 95% confidence interval [CI] for AUC=0.709–0.977) for the baseline sample, 0.832 (P=0.004; 95% CI 0.700–0.964) for the second sample, and 0.707 (P=0.146; 95% CI 0.526–0.888) for the third sample. We also analyzed different cutoff values for the assay (for defining a positive result) ranging from 0.1 to 0.4 IU/mL of interferon-γ (Table 3). Increases in the cutoff value resulted in slight improvement in positive predictive value (i.e., ability of a positive test to predict spontaneous clearance) but a significant loss in negative predictive value.
Absolute interferon-γ values were also analyzed for their correlation with viral load measurements performed the same day. In 70 samples, both viral load and CMI assays were collected on the same day. In this subset, the CMV-specific interferon-γ production was inversely correlated with CMV viral load measured at the same time (Spearman's rho −0.318, P=0.007). For instance, the median viral load at the time of any positive test (≥0.2 IU/mL) was 225 copies/mL (IQR undetectable to 1865), whereas concurrently to a negative CMI testing (<0.2 IU/mL) the viral load was 2345 copies/mL (IQR 225–8350).
Because the assay measures interferon-γ production from lymphocytes, we also compared total lymphocyte counts in the spontaneous clearance group versus those with progression, and no significant difference was seen at all time points (Table 1). However, absolute interferon-γ levels were positively correlated with the lymphocyte count in 85 samples where both values were concurrently available (Spearman's rho 0.386, P<0.001).
Asymptomatic low-level CMV viremia is very common in transplant patients; making the detection of viremia in such patients is the basis of preemptive prevention strategies for CMV. However, a subset of patients will spontaneously clear viremia without the need for treatment. This has meant that optimal thresholds for use in preemptive strategies have not been well defined, and the clinical dilemma of what to do in asymptomatic patients with low-level CMV viremia remains. This study demonstrates that a significant refinement of preemptive protocols could be made with the addition of CMI testing. We show that low-level viremia patients who have a detectable CMV-CMI response have a high likelihood of spontaneous clearance of viremia, whereas those with a negative CMI response have a high risk of virologic or clinical progression. In the clinical setting therefore, the use of CMI testing could help decide which patients with low-level viremia could simply be followed closely and which patients should commence antiviral treatment. This could result in more rational antiviral use, with a potential decrease in toxicity and cost. Although it could also be argued that treatment of any level of viremia may be important to prevent indirect effects of CMV, the true indirect effects of low-level CMV replication have yet to be ascertained. It should also be noted that prediction is not absolute and that some patients with detectable CMI responses and low-level viremia will still have virologic or clinical progression as noted for two patients in our study.
The Quantiferon-CMV assay has only been assessed in a small number of studies. Walker et al. (5) evaluated this assay in 25 heart and/or lung patients, in whom the test differentiated seronegative from seropositive transplant recipients at different times posttransplant, regardless of antiviral or immunosuppressive therapy. The association between CMV-CMI and viremia was not addressed in that study. In a previous study of 108 high-risk transplant patients, CMV CMI was assessed at the end of prophylaxis to predict late-onset CMV disease (4). Patients with a positive CMI test at the end of prophylaxis had a late-onset disease rate of 5.3% vs. 22.9% in patients with a negative CMI test. Westall et al. (6) evaluated this assay longitudinally in 39 lung transplant recipients and correlated results with CMV viral loads within the bronchoalveolar lavage fluid. Although a marked decrease in the CMI response was seen at the time of viral reactivation in the lung, the assay was not particularly predictive of significant reactivation in that organ compartment. In one additional study, this assay was evaluated in a cohort of 14 viremic kidney transplant recipients (7). A lower, although not statistically significant, interferon-γ response in patients with high viral loads compared with those with low viral loads was observed. The other clinical scenario where CMI testing might be used is to guide the duration of treatment for CMV disease and to prevent recurrent CMV disease. We are not aware of any specific studies that answer this question. Finally, another area where a CMI test would be used is to guide what to do in patients with low-level viremia (i.e., to refine a preemptive algorithm). We believe that this is the first study to specifically address this question in a rigorous manner.
Several studies have assessed intracellular cytokine staining and ELISPOT for assessment of CMV CMI (8 – 19), although the exact clinical scenarios where these tests should be applied have been more difficult to interpret. Development of CMV CMI by CD4+ and CD8+ intracellular cytokine staining was associated with control of subsequent viremia episodes without development of CMV disease in kidney and liver D+/R− patients (20). Recovery of CMV-specific response as measured by ELISPOT in a cohort of small bowel/multivisceral transplant recipients early posttransplant was protective against development of moderate or severe episodes of CMV disease (21). On the other hand, viral replication was associated with significantly lower frequencies of CMV-specific CD8+ cells measured by flow cytometry in seropositive kidney recipients yet the assay performed poorly on predicting concurrent and future CMV replication (12). Overall, these studies help confirm the importance of CD4+ and CD8+ T cell in control of viral replication posttransplant. These measurements are still not in routine use in clinical practice for a number of reasons. Flow cytometry allows assessment of both CD4+ and CD8+ T-cell populations, is quantitative, and allows measurement of cytokines other than interferon-γ. ELISPOT assays have been reported in several studies and also seem to be useful for measurement of CMV-specific immune response. For both methods complexity, cost and lack of standardization are important issues to consider. The ELISA-based Quantiferon-CMV assay is approved for in vitro diagnostics in Europe, Australia, and Korea and for investigational use only in the United States and Canada. It also uses equipment that is readily available in most laboratories. It measures primarily CD8+ T-cell responses (not CD4+) and is HLA-restricted such that patients with uncommon HLA types may not be represented in the peptide pool used for stimulation (7, 22). This could be a cause for false-negative tests.
In this study, some patients had a change in test result over subsequent measurements. A small number of patients who were initially positive became negative. This may be due to interferon-γ values near the threshold for positivity or due to subtle changes in cell mediated immune response possibly as a consequence of viral replication. More commonly, however, an overall interferon-γ production increase over time occurred (Fig. 2) presumably as an appropriate immune response to viral replication. This was especially apparent in the group that had spontaneous clearance (absolute interferon-γ increased from 1.73 to 3.75 IU/mL between baseline and follow-up measure). Serial CMI testing could also be clinically explored, although our results indicate that a single sample collected early in the course of viremia had better prediction value when compared with follow-up testing. In addition, we show that the assessment of CMV-CMI in solid organ transplant during a viremia episode does not simply reflect the pretransplant serostatus of the patient, suggesting it rather reflects the functional status of CMV-specific CD8+ T cells at the time of assessment.
Our study had a number of limitations. First, the CMI assay was not truly obtained on the same day as the onset of viremia, due to the time to obtain the initial viral load results (usually 48 hr), and return of the patient for obtaining informed consent and collecting the CMI assay. Therefore, our “baseline” CMI assessment actually reflects a few days after onset of low-level viremia in most instances. However, this situation likely reflects how the assay would be used in clinical practice, where asymptomatic viremic patients could be asked to return for assessment of their CMI response. Another limitation of our study was the small sample size. This was partly due to the stringent criteria for enrolment. These criteria were necessary so that a clear clinical question could be answered from the study, which we believe we have done. Our study population was also primarily R+ patients. Viral kinetics may differ between D+/R− patients and R+ patients. However, R+ patients are the primary group for which a preemptive strategy is recommended as a reasonable option to prophylaxis (23, 24). Finally, our study analyzed a heterogeneous organ type group, with predominance of kidney recipients. The overall results are consistent across organs. We believe that the next step would be to perform a larger multicenter validation trial.
In summary, we show that in organ transplant patients with asymptomatic low-level CMV viremia, the measurement of CMV-specific CMI using the Quantiferon-CMV assay can help predict which patients will spontaneously clear virus versus which patients will have progression. This could be used in the clinical setting as an adjunctive tool in a preemptive protocol to allow more rational and targeted antiviral use. Further studies could compare the efficacy and cost effectiveness of preemptive strategies using virologic monitoring alone versus preemptive strategies using a combination of virologic and immunologic monitoring to help guide antivirals.
MATERIALS AND METHODS
The study was approved by the institutional ethics board. Informed written consent was obtained from every patient and/or their competent representative. Adult solid organ transplant recipients with recent onset (i.e., within 1 week of first detection) low-level CMV viremia were eligible if (1) they had no symptoms of CMV disease, (2) their viral load was below the threshold for preemptive therapy at our institution (<15,000 copies/mL), and (3) they were not being treated for an episode of acute rejection. At our center, CMV donor seropositive/recipient seronegative (D+/R−) patients receive 3 months of prophylaxis and then undergo weekly viral load monitoring for 8 weeks. This strategy is also used for R+ patients who are lung transplant recipients or those given thymoglobulin induction. The remaining CMV seropositive (R+) patients are on a preemptive strategy with weekly viral load monitoring for the first 12 weeks posttransplant.
Cell-Mediated Immunity Testing
The CMI test was performed using the Quantiferon-CMV assay (Cellestis International, Melbourne, Australia). This assay is based on the measurement of interferon-γ released upon stimulation of whole blood with 21 class I HLA-restricted CMV peptides encompassing the most common HLA types present in the general population (5). The peptide pools span several immunodominant CMV epitopes including pp65, IE-1, IE-2, pp50, and gB. Assays were carried out as per manufacturer's instructions. In brief, for each test, 1 mL of whole blood was collected into each of three heparinized tubes containing either coated CMV peptides (CMV tube), a positive mitogen control (MIT tube), and a negative control with no antigen (NIL tube). After collection, tubes were shaken vigorously, and were incubated for 16 to 24 hr at 37°C. After centrifugation, supernatant was harvested and interferon-γ levels were measured (in IU/mL) by use of a standard ELISA. Results of the Quantiferon-CMV assay were not available for the management of the patients. Background levels of interferon production detected in the NIL tube are subtracted from both the values yielded by peptide (CMV) tube or mitogen (MIT) tube before result interpretation. The recommended cutoff value for CMV CMI reactivity is 0.2 IU/mL of interferon-γ, but a cutoff value of 0.1 IU/mL was also evaluated based on previous data (4). In cases where both the mitogen and the CMV tube were negative, the result was counted as negative for the purpose of analysis. The assay was performed at three time points. A baseline CMI sample was collected usually within 7 days from the first detection of viremia, and follow-up samples were drawn 7 and 14 days later.
After the first detectable low-level viremia, all patients underwent once-weekly viral load monitoring for a minimum of 4 weeks. Viral load testing was performed using an in-house internally validated real-time polymerase chain reaction assay using DNA extracted from plasma samples (lower limit of detection 50 copies/mL) (25). The institutional threshold for initiation of preemptive therapy in asymptomatic patients was 15,000 copies/mL. In addition, any patient with symptoms consistent with CMV disease was started on antiviral therapy.
Definitions of CMV disease were based on the American Society of Transplantation recommendations for use in clinical studies in organ transplant patients (26). Spontaneous clearance of viremia was defined as the development of a negative viral load (<50 copies/mL) in the absence of any specific antiviral treatment. Virologic and/or clinical progression was defined if the patient developed symptoms consistent with CMV disease or if the viral load rose to a level greater than 15,000 copies/mL. Patients with progression received antiviral treatment (intravenous ganciclovir or oral valganciclovir) at standard doses corrected for renal function when necessary.
PASW Statistics 18 (SPSS, Inc., Chicago, IL) and Graph Pad Prism 4 (GraphPad Software, Inc., La Jolla, CA) were used for statistical analysis and graph generation, respectively. Fisher's exact test was used to compare categorical variables, and Mann-Whitney U test was used to compare continuous variables. Spearman's test was used to assess the correlation between two continuous variables. Receiver operator characteristics curve analysis was used to evaluate the performance of CMV-specific interferon-γ on prediction of clearance of viremia. A P value less than 0.05 was considered statistically significant (two-tailed).
The authors thank Dr. Xiao Li Pang, Ms. Jayne Fenton, Mr. Jingzhou Huang, and Ms. Min Zhao for technical assistance and Dr. Adrian Egli for critical revision of the manuscript.
1. Humar A, Limaye AP, Blumberg EA, et al.. Extended valganciclovir prophylaxis in D+/R- kidney transplant
recipients is associated with long-term reduction in cytomegalovirus
disease: Two-year results of the IMPACT study. Transplantation 2010; 90: 1427.
2. Finlen Copeland CA, Davis WA, Snyder LD, et al.. Long-term efficacy and safety of 12 months of valganciclovir prophylaxis compared with 3 months after lung transplantation: A single-center, long-term follow-up analysis from a randomized, controlled cytomegalovirus
prevention trial. J Heart Lung Transplant
2011; 30: 990.
3. Sester M, Gärtner BC, Sester U. Monitoring CMV-spezifischer T-Zellen nach Organtransplantation [Monitoring of CMV-specific T-cell levels after organ transplantation]. J Lab Med 2008; 32: 121.
4. 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
2009; 9: 1214.
5. Walker S, Fazou C, Crough T, et al.. Ex vivo monitoring of human cytomegalovirus
-specific CD8+ T-cell responses using QuantiFERON-CMV. Transpl Infect Dis 2007; 9: 165.
6. Westall GP, Mifsud NA, Kotsimbos T. Linking CMV serostatus to episodes of CMV reactivation following lung transplantation by measuring CMV-specific CD8+ T-cell immunity. Am J Transplant
2008; 8: 1749.
7. Lochmanova A, Lochman I, Tomaskova H, et al.. Quantiferon-CMV test in prediction of cytomegalovirus
infection after kidney transplantation. Transplant
Proc 2010; 42: 3574.
8. Abate D, Saldan A, Fiscon M, et al.. Evaluation of cytomegalovirus
(CMV)-specific T cell immune reconstitution revealed that baseline antiviral immunity, prophylaxis, or preemptive therapy but not antithymocyte globulin treatment contribute to CMV-specific T cell reconstitution in kidney transplant
recipients. J Infect Dis 2010; 202: 585.
9. Cantisan S, Solana R, Lara R, et al.. CD45RA expression on HCMV-specific effector memory CD8+ T cells is associated with the duration and intensity of HCMV replication after transplantation. Clin Immunol 2010; 137: 81.
10. Crough T, Fazou C, Weiss J, et al.. Symptomatic and asymptomatic viral recrudescence in solid-organ transplant
recipients and its relationship with the antigen-specific CD8(+) T-cell response. J Virol 2007; 81: 11538.
11. Cummins NW, Deziel PJ, Abraham RS, et al.. Deficiency of cytomegalovirus
(CMV)-specific CD8+ T cells in patients presenting with late-onset CMV disease several years after transplantation. Transpl Infect Dis 2009; 11: 20.
12. Egli A, Binet I, Binggeli S, et al.. Cytomegalovirus
-specific T-cell responses and viral replication in kidney transplant
recipients. J Transl Med 2008; 6: 29.
13. Krishnan A, Zhou W, Lacey SF, et al.. Programmed death-1 receptor and interleukin-10 in liver transplant
recipients at high risk for late cytomegalovirus
disease. Transpl Infect Dis 2010; 12: 363.
14. Mattes FM, Vargas A, Kopycinski J, et al.. Functional impairment of cytomegalovirus
specific CD8 T cells predicts high-level replication after renal transplantation. Am J Transplant
2008; 8: 990.
15. Nebbia G, Mattes FM, Smith C, et al.. Polyfunctional cytomegalovirus
-specific CD4+ and pp65 CD8+ T cells protect against high-level replication after liver transplantation. Am J Transplant
2008; 8: 2590.
16. Snyder LD, Medinas R, Chan C, et al.. Polyfunctional cytomegalovirus
-specific immunity in lung transplant
recipients receiving valganciclovir prophylaxis. Am J Transplant
2011; 11: 553.
17. Sund F, Lidehall AK, Claesson K, et al.. CMV-specific T-cell immunity, viral load, and clinical outcome in seropositive renal transplant
recipients: A pilot study. Clin Transplant
2010; 24: 401.
18. Ticha O, Stouracova M, Kuman M, et al.. Monitoring of CD38high expression in peripheral blood CD8+ lymphocytes in patients after kidney transplantation as a marker of cytomegalovirus
infection. Transpl Immunol 2010; 24: 50.
19. Yang R, Xia T, Xu G, et al.. Human cytomegalovirus
specific CD8(+) T lymphocytes display interferon-gamma secretion impairment in kidney transplant
recipients with pp65 antigenemia. Transplant
Proc 2008; 40: 3500.
20. Benmarzouk-Hidalgo OJ, Cisneros JM, Cordero E, et al.. Therapeutic effect of the acquisition of cytomegalovirus
-specific immune response during preemptive treatment. Transplantation 2011; 91: 927.
21. Chiereghin A, Gabrielli L, Zanfi C, et al.. Monitoring cytomegalovirus
T-cell immunity in small bowel/multivisceral transplant
Proc 2010; 42: 69.
22. Giulieri S, Manuel O. QuantiFERON-CMV assay for the assessment of cytomegalovirus cell-mediated immunity
. Expert Rev Mol Diagn 2011; 11: 17.
23. Humar A, Snydman D. Practice ASTIDCo. Cytomegalovirus
in solid organ transplant
recipients. Am J Transplant
2009; 9 Suppl 4: S78.
24. Kotton CN, Kumar D, Caliendo AM, et al.. International consensus guidelines on the management of cytomegalovirus
in solid organ transplantation. Transplantation 2010; 89: 779.
25. Pang XL, Chui L, Fenton J, et al.. Comparison of LightCycler-based PCR, COBAS amplicor CMV monitor, and pp65 antigenemia assays for quantitative measurement of cytomegalovirus
viral load in peripheral blood specimens from patients after solid organ transplantation. J Clin Microbiol 2003; 41: 3167.
26. Humar A, Michaels M; AST ID Working Group on Infectious Disease Monitoring. American Society of Transplantation recommendations for screening, monitoring and reporting of infectious complications in immunosuppression trials in recipients of organ transplantation. Am J Transplant
2006; 6: 262.
Keywords:© 2012 Lippincott Williams & Wilkins, Inc.
Cytomegalovirus; Cell-mediated immunity; Viremia; Transplant