Background
Despite remarkable advances in the diagnostic and therapeutic modalities for its management, cytomegalovirus (CMV) remains a significant cause of serious infectious complications and occasionally mortality in immunocompromised patients [
1]. Solid organ transplant recipients are at high risk for CMV infection, especially during the first 3–12 months after transplantation, because of high initial immunosuppression. Two main strategies are used to prevent CMV infection: prophylaxis of viral infections using antiviral drugs and preemptive therapy for organ recipients who develop evidence of CMV infection during routine screening [
2‐
4]. Both strategies have resulted in significant reductions of CMV infection and CMV-related mortality. Prophylaxis usually begins shortly after transplantation and continues for a finite period of time, often in the range of 3–6 months. However, such a strategy has not led to the elimination of postprophylaxis CMV infection. Moreover, this strategy has led to a higher risk of developing anti-CMV drug resistance, a higher cost of antiviral medications, and a greater risk of side effects, with many patients who are overtreated [
5,
6].
The highest risk of CMV disease involves 15–25% of organ transplant recipients who are seronegative for CMV (R-) and receive organs from seropositive donors (D+) [
7]. Thus, the management of CMV infection in kidney graft recipients has been driven by donor and recipient pretransplant serology [
8,
9]. However, the incidence of CMV infection is significant not only in high-risk recipients (D+/R-) but also in lower-risk recipients (D−/R+, D+/R+) [
4,
10‐
12]. Moreover, some high-risk patients never develop CMV infection despite not receiving any prophylaxis treatment [
2,
13]. These observations suggest that both the presence of CMV-specific IgG antibodies and prolonged immune system activation contribute to the development of CMV infection after transplantation.
We evaluated whether a specific viral T-cell response allows for the better identification of recipients who are at high risk of CMV infection after prophylaxis withdrawal.
Discussion
To date, QuantiFERON-CMV has not been well studied with regard to the risk of postprophylaxis CMV infection in kidney transplant recipients who are subjected to such a preventive modality, regardless of pretransplant serological status.
In our study, all of the patients (except one) were either R+ or received the kidney from a seropositive donor D(+), which made them at-risk for CMV infection after transplantation. In our study, all of the patients (except one) were either R(+) or received a kidney from a seropositive donor (D[+]), which made them at-risk for CMV infection after transplantation. In the R(−) group, all of the patients who developed CMV infection and CMV disease received the kidney from D(+). This tends to confirm the fact that the CMV sero-status of both recipients and donors’ is a very important predictor of post-transplant CMV infection and CMV disease.
However, a major finding of the present study was that a negative QuantiFERON-CMV result better defines the risk of postprophylaxis CMV infection than pretransplant CMV serostatus. Furthermore, negative QuantiFERON-CMV results appear to be even more predictive of the individual risk for CMV infection. Our results are consistent with the report of Kumar et al. [
15]. These authors evaluated the predictive value of the QuantiFERON assay for determining whether patients who complete a course of therapy for CMV DNAemia require secondary antiviral prophylaxis. At the end of treatment, 48% of the recipients had a negative QuantiFERON-CMV. The recurrence of CMV infection was observed in 69% of QuantiFERON-CMV-negative recipients, despite more prolonged antivirals, in contrast to 7% of recipients who had a positive QuantiFERON-CMV (
p = 0.001).
An increasing number of studies have been conducted to investigate the feasibility of immune monitoring of the CMV-specific T-cell response as a clinical marker for predicting CMV infection posttransplantation. Sester et al. analyzed CMV-specific T-cell levels in 96 solid organ transplant recipients. They reported a strong correlation between low absolute numbers of CMV-specific CD4 T cells and the frequency of infectious episodes in lung but not kidney transplant recipients [
16]. Eid et al. analyzed IFN-γ-producing CMV-specific CD4+ and CD8+ T cells at serial time-points among 44 high- and intermediate-risk kidney transplant recipients who received 3 months of valganciclovir prophylaxis. The study found no significant association between CMV-specific T cells and the time to CMV DNAemia, but the conclusions were limited by the small number of patients who experienced CMV DNAemia [
17]. Nonetheless, the relatively homogeneous reports of kidney transplant recipients suggest a strong correlation between detection of the CMV-specific effector T-cell response and the risk of posttransplantation CMV infection [
18‐
28].
The present study found that 24% of the recipients had no detectable cellular immunity to CMV in the QuantiFERON-CMV assay, despite positive pretransplant serology. This is consistent with other studies that showed that 23–50% of anti-CMV IgG pretransplant seropositive recipients had no functional T-cell response to CMV [
18,
29]. In another study, 12% of healthy IgG-seropositive subjects had undetectable cellular immunity, based on both the QuantiFERON-CMV and Enzyme-Linked ImmunoSpot tests, suggesting the inability of certain individuals to recognize the pp65 (ppUL83) stimulus peptide or the presence of the atypical or uncommon HLA haplotypes of these subjects [
30]. The high proportion of patients with discordance between CMV cellular immunity and the humoral response may also be potentially caused by high susceptibility of the mitogen-dependent response to the action of immunosuppressive drugs, such as tacrolimus and mycophenolate mofetil.
Up to 37% of recipients who were anti-CMV IgG seronegative at the time of transplantation had a positive QuantiFERON-CMV result. This was not caused by false negatives of the CMV serology test because the repetition of CMV serology 90 days posttransplantation confirmed the negative serostatus of these recipients for IgG. The discrepancy between the results of QuantiFERON-CMV and CMV IgG serology in patients before and after kidney transplantation has been reported by others. The positive agreement (73%) between the QuantiFERON-CMV results and CMV serology in hemodialysis patients is less than the positive agreement of 88–97% that was previously reported in healthy adults [
19]. In the same study, 13% of the QuantiFERON-CMV measurements revealed positive test results in seronegative dialysis patients, and 21% of CMV-seronegative hemodialysis patients had positive T-Track CMV results. In kidney transplant recipients, the frequency of detecting CMV-reactive effector T cells in CMV-seronegative transplant recipients has been reported to be up to 30% [
20]. One possible explanation is that these recipients might have been appropriately sensitized before transplantation, despite the absence of CMV IgG in their serum. Alternatively, primary CMV-specific effector T-cell responses recover quickly and effectively immediately after transplantation, thus providing sufficient protection and control of CMV replication [
20]. Further investigations are needed to address these possibilities in kidney transplant candidates. Notably, analyses that are limited to the serological measurement of specific antibodies may not provide a sufficient assessment of the absolute memory repertoire because it excludes preformed CMV-specific memory B cells that can exist in the absence of detectable circulating anti-CMV IgG antibodies [
31,
32].
One possible limitation of the present study was that we did not include three cases of indeterminate QuantiFERON-CMV results. The possible inclusion of patients with indeterminate results has been controversial. According to the manufacturer, indeterminate results are not interpretable. For the purposes of analysis, Manuel et al. classified negative and indeterminate results together as nonreactive [
33]. Notably, 37% of the organ transplant recipients in their study were treated with thymoglobulin, which is known to increase the rate of lymphocytopenia and the rate of indeterminate QuantiFERON-CMV results. In the aforementioned study, patients with indeterminate results had the highest risk of developing CMV disease, prompting the authors to suggest that indeterminate results reflect a high net state of immunosuppression. However, this possibility was not supported by other investigators who did not confirm the predictive value of indeterminate results for determining the risk of CMV infection [
34]. In the present study, the number of patients with indeterminate results was too small to allow assessment of their risk for posttransplantation CMV infection.
In previous studies, the risk of CMV infection in organ transplant recipients was associated with maintenance immunosuppression and the usage of T-cell-depleting antibodies [
7,
8]. We did not observe a correlation between maintenance immunosuppression and the occurrence of CMV infection. A possible explanation for this fact is that the majority (98%) of the patients in the present study received the same protocol (i.e., combination of tacrolimus, steroids, and mycophenolate mofetil or mycophenolate sodium as maintenance immunosuppression). Similar to previous reports, the present study confirmed that induction therapy is a risk factor for postprophylaxis CMV infection, but such a conclusion is limited by the small number of patients who present T-cell depletion [
35].
Hypogammaglobulinemia is defined as a low level of any or all of the five classes of immunoglobulins. Although the precise mechanism of the evolution of HGG posttransplantation is unclear, secondary HGG has been found to correlate with various factors, including treatment with glucocorticosteroids and mycophenolate mofetil [
36]. Other data suggest that HGG develops more frequently after kidney transplantation in patients who are being treated with mycophenolate mofetil compared with patients who are being treated with azathioprine [
37]. One of the most interesting findings of the present study was that at the time of prophylaxis discontinuation, HGG was the strongest predictor of CMV infection. Corales et al. reported the occurrence of CMV infection in five of six heart transplant recipients with severe, profound HGG. Lymphocyte subset analysis suggested a decrease in the number of CD4+ T cells, which are required to stimulate B-cell responses [
38]. Surprisingly, very sparse information is available on the association between HGG and viral infections following kidney transplantation. Broeders et al. reported that combined HGG (IgG and IgA or IgG and IgM) was associated with more frequent infectious complications, including CMV infection. Similar to our findings, isolated hypo-IgG, although frequent, has not been identified as a predisposing factor for CMV infection after kidney transplantation [
39,
40].
Conclusions
Despite the relatively small sample size in the present study, the findings have relevant clinical implications. Current clinical protocols use very few predictors of the occurrence of postprophylaxis CMV infection. The present results indicate that negative QuantiFERON-CMV results in recipients who receive antiviral prophylaxis is a strong predictor of the risk of postprophylaxis CMV infection, independent of pretransplant CMV serology.
We also found that induction therapy, lymphocytopenia, and HGG on day 90 posttransplantation were risk factors for postprophylaxis CMV infection.
The negative QuantiFERON-CMV assay results suggest an inadequate cellular response and may be used as an indication for extending antiviral prophylaxis or monitoring CMV DNAemia after prophylaxis withdrawal.