Original Article | Published:

The kinetics of torque teno virus plasma DNA load shortly after engraftment predicts the risk of high-level CMV DNAemia in allogeneic hematopoietic stem cell transplant recipients

Bone Marrow Transplantation volume 53, pages 180187 (2018) | Download Citation

Abstract

Monitoring Torque teno virus (TTV) DNA load helps to estimate the risk of opportunistic infections in solid organ transplant recipients. We investigated whether the early kinetic pattern of plasma TTV DNA load after allogeneic hematopoietic stem cell transplantation (allo-HSCT) associates with subsequent CMV and EBV DNAemia. This study included 71 allo-HSCT patients. We found that the area under the curve (AUC) for log10 TTV DNA loads quantified by days 20 and 30 after transplantation (TTV DNA load AUC20-30), was significantly lower (P=0.036) in patients who subsequently developed CMV DNAemia requiring preemptive antiviral therapy (n=17) than in those who did not (n=8) or had no CMV DNAemia (n=19). Patients displaying TTV DNA load AUC20-302.8 copies × days × mL−1 were more likely to have high-level CMV DNAemia. A trend towards a direct correlation between TTV DNA AUC20-30 and CMV-specific interferon-γ CD8+ T-cell counts by day +30 was noted (P=0.095). However, this dynamic parameter was not useful for anticipating the occurrence of either CMV recurrences (n=12) or EBV DNAemia (n=34). In summary, it may be possible to identify a subset of allo-HSCT patients at a high risk of developing high-level CMV DNAemia by analyzing the kinetics of plasma TTV DNA load early after engraftment.

Introduction

Torque teno virus (TTV), a prototypic member of the Anelloviridae family, is a circular, negative sense, single-stranded DNA virus that infects the majority of humans worldwide and is refractory to available antiviral agents.1, 2 Monitoring TTV DNA load in blood may allow physicians to infer the risk of opportunistic infections and allograft rejection events in the solid organ transplant (SOT) setting.3, 4, 5, 6, 7, 8 In fact, SOT recipients on maintenance immunosuppression with high plasma TTV DNA loads are at increased risk of virus infections, but have a decreased risk of T or B cell-mediated allograft rejection.3, 4, 5, 6, 7, 8, 9 However, the clinical value of TTV DNA load monitoring in allogeneic hematopoietic stem cell transplantation (allo-HSCT) recipients, if any, remains to be established. A handful of studies has investigated the kinetics of plasma TTV DNAemia after Allo-HSCT.10, 11, 12, 13 Collectively, these studies clearly demonstrate that TTV DNA load decreases dramatically after conditioning and that, following engraftment, it increases in parallel to absolute lymphocyte (ALC) counts. This supports the assumption that this cell subset is the major site of TTV replication14, 15, 16 and suggests that TTV DNA load may behave as an immune system reconstitution marker. We hypothesized that the overall size of plasma TTV DNA load measured shortly after engraftment may inversely associate with the risk of viral infections such as those caused by CMV and EBV, whose control is critically dependent upon the acquisition of T-cell immunocompetence.17, 18

Patients and methods

Patients

This retrospective single-center study included 71 non-consecutive patients who underwent T-cell replete allo-HSCT at the Hematology Service at the Hospital Clínico Universitario in Valencia between April 2013 and March 2016, as a curative therapy for different hematological cancers. Only adult patients (>18 years old) with plasma samples drawn at predetermined time points (see below for more details) available for TTV DNA testing were included. Of the 71 patients in this cohort, 54 had already been included in a previous study.13 The median patient age was 55 years (range 18–70 years); demographic, baseline, and post-transplant clinical data for these patients are summarized in Table 1. The study period comprised the first 120 days after transplantation. For some analyses involving EBV (incidence of recurrent EBV DNAemia), the observation period was extended up to 6 months after allo-HSCT. The study was approved by the Hospital Clínico Universitario (INCLIVA Foundation) review board and ethics committee. All the patients gave their written informed consent prior to participating in the study.

Table 1: Demographic and clinical characteristics of the patients

Plasma specimens

Cryopreserved (−80 °C) plasma specimens were retrieved (thawed for the first time) for the analyses described herein. The TTV DNA load was quantified in samples obtained at a median of 20 days (range 14–27 days), 31 days (range 25–35 days), 41 days (range 35–47 days) and 51 days (range 45–64 days) after allo-HSCT.

Plasma TTV DNA load quantitation

We quantified the TTV DNA load with a TaqMan real-time PCR kit which amplifies a highly conserved segment of the untranslated region of the viral genome, as previously described.13, 14 DNA was extracted from 200 μl of plasma with the QIAamp DNA blood Mini QIAcube kit (Qiagen, Valencia, CA, USA) using the QIAcube extraction platform (Qiagen) following the manufacturer’s instructions. PCR amplification and amplicon detection was carried out on an ABI Prism 7500 system (PE Biosystems, Foster City, CA, USA). In our experience, the limit of quantification of the assay is approximately 25 copies/mL, and the limit of detection, nearly 10 copies/mL (75% repeatability/probability). Specimens with undetectable TTV DNA loads were assigned a value of 0 for analysis purposes. All samples from each patient were assayed simultaneously in singlets.

Management of CMV and EBV infection

CMV DNA load monitoring in plasma was performed at least once a week using a RealTime CMV PCR kit (Abbott Molecular Inc., Des Plaines, IL, USA), as previously described.18, 19 The limits of detection and quantification of this assay are~20 copies/mL (95% of probability).19 Preemptive antiviral therapy, either with valganciclovir (most patients) or foscarnet, was initiated upon detection of>1000 copies/mL (1500 IU/mL) or when the CMV DNA doubling time (dt) was2.0 days, whichever came first, as previously reported.20, 21, 22

EBV DNAemia was monitored once a week using the artus EBV PCR kit (distributed by Abbott Molecular Inc.). According to the manufacturer, the assay limit of detection is 108 copies/mL (40 IU/mL) and its quantitative range is up to 8.3 log10 IU/mL.23 Nevertheless, lower EBV DNA loads can be measured with this assay (75% of probability), and we considered these to be true values. Pre-emptive therapy with rituximab was initiated upon documentation of two consecutive plasma EBV DNA loads exceeding 1, 00 copies/mL. In some instances the decision to administer rituximab therapy was justified not only on the above criterion, but also at the attending physician’s discretion based on clinical grounds, i.e. co-presentation of acute GvHD (aGvHD).

CMV-specific T-cell immunity

CMV-specific interferon-γ-producing CD8+ T cells in whole blood were enumerated by flow cytometry to detect intracellular cytokine staining (BD FastImmune, Becton, Dickinson and Company, San Jose, CA, USA), as previously detailed.24, 25 Two sets of overlapping 15-mer peptides encompassing the entire sequence of the CMV pp65 and IE-1 proteins were simultaneously used for intracellular cytokine staining stimulation.

Definitions

CMV and EBV DNAemia were defined by the detection of CMV or EBV DNA (at any level) in one or more plasma specimens. High-level CMV DNAemia was arbitrarily defined as >1000 copies/mL (prompting the initiation of preemptive antiviral therapy at our center). The overall duration of a given episode of viral DNAemia was the time between the day of first detection of viral DNA in plasma and that of the first negative (undetectable) PCR result; aGvHD was diagnosed and graded as previously reported.26 Engraftment was defined as an absolute neutrophil count of 0.5 × 109/L, and a platelet count of 20 × 109/L.27 HLA mismatch was defined as the presence of at least one disparity in the 10 HLA sites at the A, B, C or DRB1 locus, as determined by high-resolution genotyping, both in siblings or unrelated donors.28

Statistical analysis

The cumulative incidence of CMV and EBV DNAemia was calculated using the cumulative incidence method (marginal probability) with the statistical software R (http://www.r-project.org/). Death and relapse of the underlying disease were considered as competitive events. The TTV DNA load area under the curve (AUC) was calculated with the STATGRAPHIC Centurion XVII statistics package (Statpoint Technologies, Inc., Warrenton, VA, USA). Differences between the means and medians were compared using the Student t-test or the Mann–Whitney U-test, respectively. Correlations between variables were assessed using the Pearson’s correlation test or the Spearman’s rank test, depending upon the distribution of the dataset (parametric or non-parametric, respectively). Univariate and multivariate logistic regressions were used to identify risk factors for certain clinical events. Two-sided exact P-values are reported; a P-value0.05 was considered to be statistically significant.

Results

A total of 71 patients with 4 plasma specimens collected at around days 20, 30, 40 and 50 after allo-HSCT were included in this study. The study period was defined as the first to fourth months after transplantation because most episodes of CMV and EBV DNAemia occur within this time window,17, 18 and so, following local guidelines, frequent (once a week) and systematic CMV and EBV DNA load monitoring is routinely performed during this time period.

CMV DNAemia developed in 52 patients at a median of 24.5 days after transplantation range −7 to 103 days), with a cumulative incidence of 73.3% (95% confidence interval 66.0–79.3%) within the first 120 days after allo-HSCT. Twenty-seven of these patients (51.9%) developed high-level DNAemia and were treated preemptively with antivirals. All episodes of CMV DNAemia eventually cleared, either with or without antiviral treatment. Twelve patients (23%) had a recurrent (second) episode of CMV DNAemia at a median of 73 days after transplant (range 32–117 days); no patient developed CMV end-organ disease.

EBV DNAemia was detected in 34 of the 71 patients, at a median of 79 days after transplant (range 0–175 days); its cumulative incidence was 49.1 (95% confidence interval 36.4–60.7%). Five out of these 34 patients were treated with rituximab (three patients had two consecutive EBV DNA load measurements >1000 copies/mL and the remaining two patients had grade III aGvHD and were treated at first detection of >1000 EBV DNA copies/mL); all episodes eventually cleared within the observation period. Eleven patients (32.4%) had recurrent episodes (second episode) of EBV DNAemia within the first 6 months after transplantation (median time, 135 days; range day 100–160). No patient developed post-transplant lymphoproliferative disorders.

Kinetics of plasma TTV DNA load, and incidence and features of CMV and EBV DNAemia episodes

Pre-transplant specimens were available from 45 out of the 71 patients, of which 39 had a quantifiable TTV DNA load. Overall, the mean TTV DNA load in these specimens was 2.33 log10 copies/mL (range 0–7.97 log10 copies/mL). Eventually, all patients had at least one specimen that tested positive for TTV DNA by real-time PCR. Most specimens testing positive had TTV DNA loads above the limit of quantitation of the assay. Samples testing ‘undetectable’ (<10 copies/mL) were assigned a TTV DNA load value of 0. There was a correlation between pre-transplant TTV DNA loads and those measured by day 20 (P=0.614; P=0.001) and 30 (P=0.354; P=0.017) after transplantation.

We hypothesized that the magnitude of the AUC for log10 TTV DNA loads, quantified between days 20 and 30 (TTV DNA load AUC20-30) after transplantation (median time to engraftment, 16 days) inversely associated with the risk of subsequent CMV or EBV DNAemia occurrence. It is of interest that the TTV DNA load AUC20-30 was comparable, irrespective of the type of transplant and conditioning, HLA-matching, D/R CMV serostatus and the use of cyclophosphamide (post transplant) or anti-thymocyte globulin in the conditioning regimen (Supplementary Table 1). The small number of allografts other than peripheral blood (n=2) precluded any meaningful analysis on the impact of the stem cell source on TTV DNA load AUC20-30.

We observed that the mean TTV DNA load AUCs20-30 was lower in patients with subsequent CMV DNAemia (3.3 copies × days × mL−1; range 0–7.62) than in patients without it (4.4 copies × day × mL−1; range 0–8.43), although the difference did not reach statistical significance (Figure 1a). Nevertheless, the mean TTV DNA load AUC20-30 was significantly lower (Figure 1b) in patients with subsequent high-level CMV DNAemia who were eventually treated preemptively with antivirals (n=17) (2.7 copies × days × mL−1; range 0–7.62) than in those who were not treated (n=8) or had no documented CMV DNAemia (n=19) (4.4 copies × days × mL−1; range 0–8.43). For the above analyses, 27 patients with CMV DNAemia occurring before day 30 were excluded.

Figure 1
Figure 1

AUC for log10 TTV DNA loads quantified between days 20 and 30 after transplant (TTV DNA load AUC20-30) in patients with or without CMV DNAemia developing after day 30 post transplant (a) and in patients with or without CMV DNAemia episodes after day 30 post-transplant requiring preemptive antiviral therapy, as decided according to local guidelines (b). Bars indicating the mean values and the standard deviations as well as P-values for comparisons between groups (Student’s t-tests) are shown.

As can be observed in Figure 1a, seven patients in the treatment group had TTV DNA load AUC20-30s=0. Five out of these patients had quantifiable TTV DNA loads by day 40 and the remaining 2 by had it day 60 after transplant. The ALC increase between days 20 and 30 was lower in these patients than that in patients with TTV DNA load AUC20-30s >0 (not shown). None of these patients had been treated with anti-thymocyte globulin and were clinically heterogeneous (Supplementary Table 2).

The TTV DNA load AUC20-30 was not significantly correlated with the duration of CMV DNAemia episodes (median, 51 days; range 18–153) and the peak CMV DNA level within episodes (median, 3.43 log10/mL; range 1.72–6.43), although an inverse trend was noted (Figures 2a and b, respectively).

Figure 2
Figure 2

Correlation between the AUC for log10 TTV DNA loads quantified between days 20 and 30 after transplant (TTV DNA load AUC20-30) and the duration of CMV DNAemia (a) and the plasma peak CMV DNA load (b) during initial (first) episodes. Correlation between these variables was analyzed using the Pearson’s test. R and P-values are shown.

Of note, the incidence of grades II-IV aGvHD, known to increase TTV DNA load levels,12, 13 before day 30 post-allo-HSCT and throughout the study, was comparable in patients with or without CMV DNAemia, whether or not they required preemptive antiviral therapy (P=0.363 and P=0.714, respectively).

A ROC curve was built to define the TTV DNA load AUC20-30 that would best discriminate between patients with or without subsequent high-level CMV DNAemia (Supplementary Figure 1). This value was found to be 2.8 copies × days × mL−1. As can be inferred from Figure 1b, 11 patients had values 2.8, of whom 7 developed high-level CMV DNAemia (predictive value for this event, 64%); conversely, 23 out of 33 patients with TTV DNA load AUC20-30 >2.8 did not develop high-level CMV DNAemia (predictive value for protection, 70%). This threshold value was entered into univariate models and found to be significantly associated with an increased risk of high-level CMV DNAemia, along with the receipt of an allograft from a CMV-seronegative donor (for CMV-seropositive recipients). Both factors, nevertheless, lost significance in multivariate analyses (Table 2).

Table 2: Risk factors for CMV DNAemia requiring preemptive antiviral therapy

In contrast to the TTV DNA load AUC 20-30, the ALC log10 count AUC20-30 was not associated with the occurrence of high-level CMV DNAemia (P=0.156), although lower values were observed in these patients (mean, 3.4 cells × days × μL−1 in patients with high-level DNAemia vs 4.43 cells × days × μL−1 in patients without it; P=0.156).

Likewise, neither pre-transplant TTV DNA load nor those quantified by day 20 and by day 30 after allo-HSCT in patients with subsequent high-level CMV DNAemia did differ significantly from that in patients without it, although higher levels were seen in the latter group at days 20 and 30 after transplantation (Table 3).

Table 3: Pre-transplant and post-transplant plasma TTV DNA loads and occurrence of CMV DNAemia in allo-HSCT recipients

The incidence of recurrent episodes of CMV DNAemia (second episode) was also not related to the TTV DNA load AUC20-30, (P=0.247; Supplementary Figure 2).

To investigate the relationship between the kinetics of TTV DNA load and the risk of EBV DNAemia we considered the AUCs between day 20 and day 50 after transplantation (TTV DNA load20-50), in addition to the TTV DNA load AUC20-30. This is because most EBV DNAemia episodes occurred after day 50; 8 and 4 patients were excluded from these analyses, because they presented EBV DNAemia before days 50 and 30 post transplant, respectively. The TTV DNA load AUC20-30 was comparable in patients with (3.93 copies × daysmL−1μL−1; range 0–7.62) or without (4.47 copies × days × mL−1; range 0–9.85) subsequent EBV DNAemia (Figure 3a). Likewise, the TTV DNA load AUCs20-50 (Figure 3b) were not significantly different between groups (6.67 copies × days × mL−1; range 4.18–8.62 in patients with EBV DNAemia and 7.10 copies × days × mL−1; range 0–9.90 in patients without it). In addition, neither TTV DNA load AUC20-30 nor TTV DNA load AUC20-50 were correlated with the risk of recurrent EBV DNAemia within the first 6 months after transplantation (P=>0.5). We found no association between either the TTV DNA load AUC20-30 or AUC20-50 and the occurrence of EBV DNAemia episodes requiring rituximab treatment (P=0.52 and P=0.98, respectively).

Figure 3
Figure 3

AUC for log10 TTV DNA loads quantified between days 20 and 30 (TTV DNA load AUC20-30) (a) and between days 20 and 50 (TTV DNA load AUC20-50) (b) in patients with or without EBV DNAemia developing after days 30 and 50 post transplant, respectively. Bars indicating the mean values and the s.d., as well as P-values for comparisons between groups (Student’s t-tests) are shown.

Kinetics of plasma TTV DNA load and early reconstitution of CMV-specific T-cell immunity

We next investigated whether TTV DNA load AUC20-30 did correlate with CMV-specific interferon-γ CD8+ T-cell counts enumerated by day +30 after allo-HSCT. Data on CMV-specific T-cell immunity were available for 21 patients who had not previously presented CMV DNAemia. We noted a non-statistically significant trend towards a direct correlation between TTV DNA AUC20-30 and peripheral counts of this T-cell subset (P=0.095; Figure 4). Of interest, median CMV-specific interferon-γ CD8+ T-cell counts were comparable in D-/R and D+/R+ pairs (P=0.470).

Figure 4
Figure 4

Correlation between the AUC for log10 TTV DNA loads quantified between days 20 and 30 after transplant (TTV DNA load AUC20-30) and peripheral levels of CMV pp65/IE-1-specific interferon-γ-producing CD8+ T cells measured at day 30 after transplant. Correlation between these variables was analyzed using the Spearman rank test. ρ and P-values are shown.

Interaction between CMV and TTV

To investigate whether the occurrence of CMV DNAemia had any impact on the dynamics of plasma TTV DNA load we compared the difference in TTV DNA loads between day 40 and day 20, and between day 50 and day 20 after allo-HSCT in patients in whom CMV DNAemia did (n=18 and n=21, respectively) or did not develop (n=19) within these time windows. Patients who had CMV DNAemia, either prior to day +20 or after day +40 or +50, respectively, were excluded from the analyses. Our results indicated that CMV DNAemia had no apparent effect on the kinetics of TTV DNA load (P=0.461 and P=0.748, respectively). In support of this view, as shown in Supplementary Table 1, the TTV DNA load AUC20-30 was comparable (P=0.195) in patients with CMV DNAemia developing prior to day 30 after allo-HSCT (mean, 4.5 copies × days × mL−1; range 0–9.8) and in patients with no CMV DNAemia within this period of time (mean 3.7 copies × days × mL−1) Moreover, no overall correlation was found between TTV DNA and CMV DNA loads over the study period, regardless of whether all plasma specimens or only those yielding quantifiable TTV DNA loads for both viruses were considered for the analyses (Figures 5a and b, respectively).

Figure 5
Figure 5

Correlation between the TTV and CMV DNA loads in plasma samples testing positive for both viruses by real-time PCR (a) or in all specimens, including those testing negative for TTV, CMV or both (b). Correlation between these variables was analyzed using the Spearman’s rank test. ρ and P-values are shown.

Discussion

This is the first study to suggest that the overall TTV DNA load quantified early after engraftment, between days 20 and 30 after allo-HSCT (TTV DNA load AUC20-30), may be an ancillary parameter to predict the occurrence of subsequent high-level plasma CMV DNAemia. The information derived from this parameter could thus complement that obtained by analyzing the CMV-specific functional T-cell response or the kinetics of CMV DNA load in the blood compartment to more precisely identify patients at risk of developing high-level CMV replication episodes.17 In the setting of preemptive antiviral therapy, early inception of treatment to avoid exceedingly high CMV loads may improve patient’s survival.28, 29

Although the size of the TTV DNA load AUC20-30 was not significantly different between patients with or without subsequent CMV DNAemia, a trend towards an inverse relationship between these two parameters was evident. Nevertheless, patients who developed episodes of high-level DNAemia that eventually required the inception of preemptive antiviral treatment displayed a significantly lower mean TTV DNA AUCs20-30 than those who were able to clear the infection without the need for therapy or those with no documented CMV DNAemia. This was the case irrespective of the criterion used to trigger the initiation of therapy: either a CMV DNA doubling time2 days (n=2) or a CMV DNA load>1000 copies/mL (n=15). It should be underlined here that the above plasma CMV DNA threshold is commonly applied for triggering the administration of preemptive therapy in this setting.17

Moreover, we identified a TTV DNA load AUC20-30 threshold (2.8 copies × days × mL−1) that was associated with an increased risk of high-level CMV DNAemia episodes requiring preemptive antiviral treatment in univariate models, even though it displayed a rather modest predictive value for this event. Likewise, D−/R+ CMV patients were at higher risk of developing such type of episodes when compared to their D+/R+ counterparts, as previously reported.17 No other factor such as the type of allograft, the conditioning regimen, the use of anti-thymocyte globulin, post-transplant cyclophosphamide, or the occurrence of severe aGvHD was associated with high-level CMV DNAemia in this cohort.

It is worth noting that the inclusion of seven patients displaying TTV DNA load AUC20-30s=0 within the group of patients who developed high-level CMV DNAemia had a critical impact on the results. These patients showed a modest or no increase in ALCs between days 20 and 30 after transplantation (not shown), were clinically heterogeneous and had not been treated with anti-thymocyte globulin, a well known T-cell depleting agent that has a major impact on TTV DNA load kinetics in SOT.6 Nevertheless, four of these patients received a bone marrow, cord blood or an HLA-mismatched allograft; all these conditions are known to be associated with a delayed reconstitution of virus-specific T-cell immunity.17

Although the data reported herein show promise, they should be interpreted with caution, provided that only a relatively scarce number of high-level CMV DNAemia episodes (n=17) were included in the analyses. In this context, we chose not to take into consideration episodes of CMV DNAemia developing prior to day 30 after transplantation (nearly 50%), irrespective of whether or not they ultimately required the inception of antiviral treatment, because the possibility that CMV replication modulates plasma TTV DNA load kinetics could not absolutely be ruled out here, even though our data seemed to indicate that this was not the case, as discussed below.

Interestingly, despite plasma TTV DNA loads correlating with ALC following engraftment,13 we found that the magnitude of ALC AUC 20-30 was not associated significantly with the occurrence of high-level CMV DNAemia episodes; nevertheless, higher AUCs were usually seen in patients with self-resolving episodes.

Allo-HSCT patients who fail to reconstitute CMV-specific T-cell immunity early after transplant are known to be at higher risk of developing high-level viremia that eventually requires the inception of antiviral therapy for clearance.1, 24, 25 In this context, we observed that CMV pp65/IE-1-specific CD8+ T-cell counts, enumerated by day +30 after transplant in patients with no previous DNAemia, tended to correlate with TTV DNA load AUCs20-30, hence suggesting that this parameter may behave as a surrogate marker for CMV-specific T-cell reconstitution. However, this correlation did not reach statistical significance, likely because the number of patients in our sample with T-cell immunity data available was limited, thus precluding us from drawing any robust conclusions on this issue. Further supporting the above assumption, we observed a trend towards an inverse association between TTV DNA load AUC20-30 and CMV DNA peak values within episodes of CMV DNAemia.

We were unable to infer the risk of recurrent CMV DNAemia based upon TTV DNA load AUC20-30 values. This, nevertheless, was not entirely unexpected, as such a risk is largely dependent on the strength of functional CMV-specific T-cell expansion within the initial episode of virus replication.30, 31, 32

We also found no association between the incidence of initial or recurrent episodes of EBV DNAemia and the TTV DNA load AUCs20-30 and AUCs20-50. The limited number of patients receiving rituximab for EBV DNAemia in our cohort precluded any meaningful assessment of the clinical value of TTV DNA load AUCs for predicting this event. In the absence of data on EBV-specific T-cell immunity, we speculate that the time span for TTV DNA load measurements (the longest period between days 20 and 50) may have been insufficient to gauge the level of EBV-specific T-cell reconstitution, which often occurs at later stages after allo-HSCT.33, 34

Our data differ from those of Gilles et al.35 who found significantly higher TTV DNA loads at day +30 after transplantation in patients with CMV or EBV reactivation and/or aGvHD, thus supporting the idea that TTV DNA load may behave as a surrogate marker for immunosuppression. In our series, TTV DNA loads quantified prior to transplant as well as those those measured at days 20 and 30 after transplantation were comparable in patients with or without subsequent CMV DNAemia (either requiring preemptive antiviral therapy or not), although a trend towards lower TTV DNA load levels by days 20 and 30 after transplantation was observed in the former patients. Nevertheless, these two studies differ in the analytical characteristics of the real-time PCRs used for TTV, CMV and EBV detection and quantification.35 Perhaps more importantly, there are substantial differences in the clinical characteristics of the patients included in each of the comparison groups: in Gilles et al.35 all the patients with CMV or EBV reactivation appeared to have developed aGvHD, a condition which leads to a significant increase in TTV DNA load,12, 13 whereas in our study patients with or without CMV or EBV DNAemia (treated or untreated) were balanced in terms of the incidence of aGvHD.

CMV is a potent pro-inflammatory and immunosuppressive agent17 and as such it may modulate TTV DNA load in the blood compartment. In fact, CMV-seropositive healthy individuals appear to display higher plasma TTV DNA loads than their CMV-seronegative counterparts, at least within certain age ranges.36 Despite this, we found no evidence suggesting that CMV replication may impact on TTV DNA load kinetics within the first 50 days of allo-HSCT; first, the magnitude of the increase in TTV DNA load early after engraftment (between days 20 and 30 after transplantation) and the overall TTV DNA load within this time frame, captured by the TTV DNA load AUC 20-30, was apparently similar whether or not CMV DNAemia was detectable prior to day 30. Second, we found no correlation between TTV DNA and CMV DNA loads over the study period. Despite the above, and as previously stated, further studies are required to prove this assumption.

No reliable biological marker for identifying allo-HSCT patients at high-risk of high-level viremia is currently available other than functional CMV-specific T-cell levels in peripheral blood and this is better at predicting protection rather than the risk.17 Early assessment of the kinetics of TTV DNA load may perhaps help to solve this unmet need in a subset of patients. Unfortunately, patients with early-onset CMV DNAemia (either requiring antiviral therapy or not), nearly 50% in this cohort, would certainly not benefit from these analyses, at least in the way they were performed here.

On the basis of the data presented here and elsewhere,13 our current working hypothesis postulates that plasma TTV DNA load may behave as a surrogate marker of T-cell immune reconstitution (at least that preventing high-level CMV DNAemia) shortly after engraftment, just the opposite as observed in SOT recipients undergoing maintenance immunosuppression.3, 4, 5, 6, 7, 8 In our view, this can be explained on the basis of the following facts: (i) TTV mainly replicates in lymphocytes;16 (ii) cytopenia (lymphopenia) ensues following conditioning; immunosuppresion induction regimens in SOT do not cause such a profound lymphopenia (unless T-cell depleting agents are used at high doses); (iii) rapid lymphocyte repopulation following engraftment associates with a decreased risk of opportunistic infections;17 and (iv) a direct correlation exists between ALCs and TTV DNA load shortly after engraftment.13

Nevertheless, the occurrence of severe aGvHD requiring early after allo-HSCT, requiring the use of high-dose corticosteroids, known to increase plasma TTV DNA load,12, 13 may blur this apparent relationship. It remains to be determined whether at later times, once ALC recover and plasma TTV DNA load reaches a plateau (this time point may perhaps vary depending upon the underlying disease, the immunosuppression regimen, the type of allograft), the magnitude of this parameter reflects instead the net state of immunosuppression, as it does in SOT. Studies addressing this issue are currently underway.

The main limitations of this study are its retrospective design, which made it impossible to collect a large number of specimens for the CMV-specific T-cell immunity analyses and, again, the relatively small size of the cohort. Nevertheless, given the potential clinical relevance of our findings, prospective and well-powered studies, which can validate or refute our findings are warranted.

References

  1. 1.

    . History of discoveries and pathogenicity of TT viruses. Curr Top Microbiol Immunol 2009; 331: 1–20.

  2. 2.

    , , , . Torquetenovirus: the human virome from bench to bedside. Clin Microbiol Infect 2016; 22: 589–593.

  3. 3.

    , , , , , et al. Temporal response of the human virome to immunosuppression and antiviral therapy. Cell 2013; 155: 1178–1187.

  4. 4.

    , , , , , . Pre-transplant plasma Torque Teno virus load and increase dynamics after lung transplantation. PLoS ONE 2015; 10: e0122975.

  5. 5.

    , , , , , et al. Torque Teno Virus in children who underwent orthotopic liver transplantation: new insights about a common pathogen. J Infect Dis 2014; 209: 247–254.

  6. 6.

    , , , . Torquetenovirus (TTV) viremia correlates with intensity of maintenance immunosuppression in adult orthotopic liver transplantation. J Infect Dis 2014; 210: 667–668.

  7. 7.

    , , , , . Plasma DNA levels of Torque teno virus and immunosuppression after lung transplantation. J Heart Lung Transpl 2014; 33: 320–328.

  8. 8.

    , , , , , et al. Torque Teno virus load-inverse association with antibody-mediated rejection after kidney transplantation. Transplantation 2017; 101: 360–367.

  9. 9.

    , , , , , et al. Torque Teno Virus load and acute rejection after orthotopic liver transplantation. Transplantation 2017; 101: e219–e221.

  10. 10.

    , , , , , et al. TT virus in bone marrow transplant recipients. Blood 1999; 93: 2485–2490.

  11. 11.

    , , , , , et al. Role of hematopoietic cells in the maintenance of chronic human torquetenovirus plasma viremia. J Virol 2010; 84: 6891–6893.

  12. 12.

    , , , , , . Torque teno virus in patients undergoing allogeneic hematopoietic stem cell transplantation for hematological malignancies. Bone Marrow Transplant 2016; 51: 440–442.

  13. 13.

    , , , , , et al. Dynamics of Torque Teno virus plasma DNAemia in allogeneic stem cell transplant recipients. J Clin Virol 2017; 94: 22–28.

  14. 14.

    , , ,, , , et al. TT virus (TTV) loads associated with different peripheral blood cell types and evidence for TTV replication in activated mononuclear cells. J Med Virol 2001; 64: 190–194.

  15. 15.

    , , , , , et al. TT virus replicates in stimulated but not in nonstimulated peripheral blood mononuclear cells. Virology 2002; 301: 121–129.

  16. 16.

    , , , , . Short-term kinetics of torque teno virus viraemia after induction immunosuppression confirm T lymphocytes as the main replication-competent cells. J Gen Virol 2015; 96: 115–117.

  17. 17.

    , , , , . An update on the management and prevention of cytomegalovirus infection following allogeneic hematopoietic stem cell transplantation. Future Virol 2015; 10: 113–134.

  18. 18.

    , , , , , et al. Risk factors for lymphoproliferative disorders after allogeneic hematopoietic cell transplantation. Blood 2009; 113: 4992–5001.

  19. 19.

    , , , , , et al. Comparison of the new Abbott real time CMV assay and the Abbott CMV PCR Kit for the quantitation of plasma cytomegalovirus DNAemia. Diagn Microbiol Infect Dis 2013; 75: 207–209.

  20. 20.

    , , , , , et al. Pre-emptive antiviral therapy for active CMV infection in adult allo-SCT patients guided by plasma CMV DNAemia quantitation using a real-time PCR assay: clinical experience at a single center. Bone Marrow Transplant 2013; 48: 1010–1012.

  21. 21.

    , , , , . Early kinetics of plasma cytomegalovirus DNA load in allogeneic stem cell transplant recipients in the era of highly sensitive real-time PCR assays: does it have any clinical value? J Clin Microbiol 2014; 52: 654–656.

  22. 22.

    , , , , , et al. Preemptive antiviral therapy for CMV infection in allogeneic stem cell transplant recipients guided by the viral doubling time in the blood. Bone Marrow Transplant 2016; 51: 718–721.

  23. 23.

    , , , . Comparison of the artus Epstein-Barr virus (EBV) PCR kit and the Abbott RealTime EBV assay for measuring plasma EBV DNA loads in allogeneic stem cell transplant recipients. Diagn Microbiol Infect Dis 2017; 88: 36–38.

  24. 24.

    , , , , , et al. Enumeration of cytomegalovirus-specific interferongamma CD8+ and CD4+ T cells early after allogeneic stem cell transplantation may identify patients at risk of active cytomegalovirus infection. Haematologica 2008; 93: 1434–1436.

  25. 25.

    , , , , , et al. Reconstitution of CMV pp65 and IE-1-specific IFN-γ CD8(+) and CD4(+) T-cell responses affording protection from CMV DNAemia following allogeneic hematopoietic SCT. Bone Marrow Transplant 2011; 46: 1437–1443.

  26. 26.

    , , , , , et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors. Transplantation 1974; 18: 295–304.

  27. 27.

    , , , , , et al. Assessment of hematologic progenitor engraftment by complete reticulocyte maturation parameters after autologous and allogeneic hematopoietic stem cell transplantation. Haematologica 2001; 86: 24–29.

  28. 28.

    , , , , , et al. Impact of cytomegalovirus DNAemia on overall and non-relapse mortality in allogeneic stem cell transplant recipients. Transpl Infect Dis 2017; 19: doi: 10.1111/tid.12717.

  29. 29.

    , , , , , et al. Cytomegalovirus viral load and mortality after haemopoietic stem cell transplantation in the era of pre-emptive therapy: a retrospective cohort study. Lancet Haematol 2016; 3: e119–e127.

  30. 30.

    , , , , , et al. Lack of prompt expansion of cytomegalovirus pp65 and IE-1-specific IFNgamma CD8+ and CD4+ T cells is associated with rising levels of pp65 antigenemia and DNAemia during pre-emptive therapy in allogeneic hematopoietic stem cell transplant recipients. Bone Marrow Transplant 2010; 45: 543–549.

  31. 31.

    , , , , , et al. Immunological monitoring for guidance of preemptive antiviral therapy for active cytomegalovirus infection in allogeneic stem-cell transplant recipients: a pilot experience. Transplantation 2011; 92: e17–e20.

  32. 32.

    , , , , , et al. Efficacy and safety of a preemptive antiviral therapy strategy based on combined virological and immunological monitoring for active cytomegalovirus infection in allogeneic stem cell transplant recipients. Open Forum Infect Dis 2016; 3: ofw107.

  33. 33.

    , , , , , et al. Rapid reconstitution of Epstein-Barr virus-specific T lymphocytes following allogeneic stem cell transplantation. Blood 2000; 96: 2814–2821.

  34. 34.

    , , , , , et al. Impaired recovery of Epstein-Barr virus (EBV)—specific CD8+ T lymphocytes after partially T-depleted allogeneic stem cell transplantation may identify patients at very high risk for progressive EBV reactivation and lymphoproliferative disease. Blood 2003; 101: 4290–4297.

  35. 35.

    , , , , , et al. Dynamics of Torque Teno virus viremia could predict risk of complications after allogeneic hematopoietic stem cell transplantation. Med Microbiol Immunol 2017; doi: 10.1007/s00430-017-0511-4.

  36. 36.

    , , , , , . TTV DNA plasma load and its association with age, gender, and HCMV IgG serostatus in healthy adults. Age (Dordr) 2014; 36: 9716.

Download references

Acknowledgements

We thank Juan Manuel Moya Serrano for his technical assistance. Estela Giménez holds a Río-Hortega Contract from the Carlos III Health Institute (ISCIII) (Ref. CM16/00200). This work was presented, in part, at the European Congress of Clinical Virology in Lisbon, 2016.

Author information

Affiliations

  1. Microbiology Service, Hospital Clínico Universitario, Institute for Research INCLIVA, Valencia, Spain

    • E Albert
    • , E Giménez
    •  & D Navarro
  2. Hematology Service, Hospital Clínico Universitario, Institute for Research INCLIVA, Valencia, Spain

    • C Solano
    • , A Pérez
    • , J L Piñana
    •  & J C H Boluda
  3. Department of Medicine, School of Medicine, University of Valencia, Valencia, Spain

    • C Solano
  4. North-Western Tuscany Blood Bank, Pisa University Hospital, Pisa, Italy

    • D Focosi
    •  & F Maggi
  5. Virology Unit, Pisa University Hospital, Pisa, Italy

    • L Macera
  6. Department of Microbiology, School of Medicine, University of Valencia, Valencia, Spain

    • D Navarro

Authors

  1. Search for E Albert in:

  2. Search for C Solano in:

  3. Search for E Giménez in:

  4. Search for D Focosi in:

  5. Search for A Pérez in:

  6. Search for L Macera in:

  7. Search for J L Piñana in:

  8. Search for J C H Boluda in:

  9. Search for F Maggi in:

  10. Search for D Navarro in:

Competing interests

The authors declare no conflict of interest.

Corresponding author

Correspondence to D Navarro.

Supplementary information

About this article

Publication history

Received

Revised

Accepted

Published

DOI

https://doi.org/10.1038/bmt.2017.235

Supplementary Information accompanies this paper on Bone Marrow Transplantation website (http://www.nature.com/bmt)