Torque teno virus viremia and QuantiFERON®-CMV assay in prediction of cytomegalovirus reactivation in R+ kidney transplant recipients

Abstract

Introduction: Cytomegalovirus (CMV) is the most frequent infectious complication following solid organ transplantation. Torque teno viruses (TTV) viremia has been proposed as a biomarker of functional immunity in the management of kidney transplant recipients (KTR). The QuantiFERON^®-CMV (QF-CMV) is a commercially available assay that allows the assessment of CD8⁺ T-cell responses in routine diagnostic laboratories.

Methods: In a prospective national multicenter cohort of 64 CMV-seropositive (R+) KTR, we analyzed the value of TTV load and the two markers of the QF-CMV assay [QF-Ag (CMV-specific T-cell responses) and QF-Mg (overall T-cell responses)], alone and in combination, in prediction of CMV reactivation (≥3 log₁₀ IU/ ml) in the first post-transplant year. We compared previously published cut-offs and specific cut-offs optimized from ROC curves for our population.

Results: Using the conventional cut-off (3.45 log₁₀ copies/ml), TTV load at D0 [inclusion visit on the day of transplantation before induction (D0)], or at M1 (1-month post-transplant visit) perform better in predicting CMV viremia control than CMV reactivation. Survival analyses suggest a better performance of our optimized TTV cut-offs (3.78 log₁₀ copies/ml at D0 and 4.23 log₁₀ copies/ml at M1) for risk stratification of CMV reactivation in our R+ KTR cohort. The QF-CMV (QF-Ag = 0.2 IU/ml, and QF-Mg = 0.5 IU/ml) also appears to better predict CMV viremia control than CMV reactivation. Moreover, survival analyses suggest that the QF-Mg would perform better than the QF-Ag in stratifying the risk of CMV reactivation. The use of our optimized QF-Mg cut-off (1.27 IU/ml) at M1 further improved risk stratification of CMV reactivation. Using conventional cut-offs, the combination of TTV load and QF-Ag or TTV load and QF-Mg did not improve prediction of CMV viremia control compared to separate analysis of each marker but resulted in an increase of positive predictive values. The use of our cut-offs slightly improved risk prediction of CMV reactivation.

Conclusion: The combination of TTV load and QF-Ag or TTV load and QF-Mg could be useful in stratifying the risk of CMV reactivation in R+ KTR during the first post-transplant year and thereby have an impact on the duration of prophylaxis in these patients.

Clinical trial registration: ClinicalTrials.gov registry, identifier NCT02064699.

Keywords: CMV-seropositive recipients; QuantiFERON® CMV; Torquetenovirus; cytomegalovirus; kidney transplantation.

Figure 1

TTV load dynamics measured at transplantation (D0) and at month (M) 1, 2, 3, 4, 5, 6, 8, 10, and 12 after transplantation. The results are presented in the overall study population (A), and in the subgroups of patients with (light grey) and without prophylaxis (dark grey) at D0 (B). Mean values are displayed on the graph as points. Median values are represented by bars, and the interquartile range is represented by boxes. Significant differences are represented in the figure by an asterisk (p < 0.05).

Figure 2

TTV load dynamics of all R+ KTR according to CMV reactivation. TTV load was measured at transplantation (D0) and at month (M) 1, 2, 3, 4, 5, 6, 8, 10, and 12 after transplantation. TTV kinetics are presented in the subgroups of patients with CMV reactivation ≥3 log₁₀ IU/ml (dark grey) and without CMV reactivation (light gray) in the year post-transplant. Mean values are displayed on the graph as points. Median values are represented by bars, and the interquartile range is represented by boxes. Significant differences are represented in the figure by an asterisk (p < 0.05).

Figure 3

Kaplan–Meier survival curves modelling freedom from CMV reactivation through day + 360 post-transplant according to the result of TTV load at D0 or M1. The results are presented according to the TTV cut-off of 3.45 log₁₀ copies/ml at D0 (A), the TTV cut-off of 3.45 log₁₀ copies/ml at M1 (B), the optimal TTV cut-off at D0 determined by ROC curves (3.78 log₁₀ copies/ml) (C), and the optimal TTV cut-off at M1 as defined by ROC curves (4.23 log₁₀ copies/ml) (D). Dotted lines indicate TTV load above the cut-off while solid lines indicate TTV load below the cut-off. HR, hazard ratio and CI, confidence interval.

Figure 4

Kaplan–Meier survival curves modelling freedom from CMV reactivation through day + 360 post-transplant according to the dynamic change in TTV levels between D0 and M1. The dotted line indicates an increase in TTV levels >0.75 log₁₀ copies/ml while the solid line indicates a TTV dynamic change below this value. HR, hazard ratio and CI, confidence interval.

Figure 5

QF-CMV assay dynamics measured at transplantation (D0) and at month (M) 1, 2, 3, 4, 5, 6, 8, 10, and 12 after transplantation. The results are presented separately for the QF-Ag and the QF-Mg in the overall study population (A,C), and in the subgroups of patients with (light grey) and without prophylaxis (dark grey) at D0 (B,D). Mean values are displayed on the graph as points. Median values are represented by bars, and the interquartile range is represented by boxes. Significant differences are represented in the figure by an asterisk (p < 0.05).

Figure 6

QF-CMV assay dynamics of all R+ KTR according to CMV reactivation. QF-assay was measured at transplantation (D0) and at month (M) 1, 2, 3, 4, 5, 6, 8, 10, and 12 after transplantation. QF-Ag (A) and QF-Mg (B) kinetics are presented in the subgroups of patients with CMV reactivation ≥3 log₁₀ IU/ml (dark grey) and without CMV reactivation (light grey), as measured in the year post-transplant. Mean values are displayed on the graph as points. Median values are represented by bars, and the interquartile range is represented by boxes. Significant differences are represented in the figure by an asterisk (p < 0.05).

Figure 7

Kaplan–Meier survival curves modelling freedom from CMV reactivation from M1 to M12 according to the result of the QF-CMV at M1. The results are presented for the manufacturer’s QF-Ag cut-off (A), the optimal QF-Ag cut-off determined by ROC curves (B), the manufacturer’s QF-Mg cut-off (C), and the optimal QF-Mg cut-off determined by ROC curves (D). Solid lines indicate QF-CMV results above the cut-off while dotted lines indicate QF-CMV results below the cut-off. HR, hazard ratio and CI, confidence interval.