Comparison of Three Cellular Assays to Predict the Course of CMV Infection in Liver Transplant Recipients

Abstract

To estimate protection from cytomegalovirus (CMV) replication after solid organ transplantation, CMV serology has been considered insufficient and thus CMV immunity is increasingly assessed by cellular in vitro methods. We compared two commercially available IFN-γ ELISpot assays (T-Track CMV and T-SPOT.CMV) and an IFN-γ ELISA (QuantiFERON-CMV). Currently, there is no study comparing these three assays. The assays were performed in 56 liver transplant recipients at the end of antiviral prophylaxis and one month thereafter. In CMV high- or intermediate-risk patients the two ELISpot assays showed significant correlation (p < 0.0001, r > 0.6) but the correlation of the ELISpot assays with QuantiFERON-CMV was weaker. Results of both ELISpot assays were similarly predictive of protection from CMV-DNAemia ≥500 copies/mL [CMV pp65 T-SPOT.CMV at the end of prophylaxis: area under curve (AUC) = 0.744, cut-off 142 spot forming units (SFU), sensitivity set to 100%, specificity 46%; CMV IE-1 T-Track CMV at month 1: AUC = 0.762, cut-off 3.5 SFU, sensitivity set to 100%, specificity 59%]. The QuantiFERON-CMV assay was inferior, reaching a specificity of 23% when setting the sensitivity to 100%. In conclusion, both CMV-specific ELISpot assays appear suitable to assess protection from CMV infection/reactivation in liver transplant recipients.

Keywords: ELISA; ELISpot; human cytomegalovirus; interferon-γ; liver transplantation; prediction; reactivation.

Figure 1

Correlation of CMV-specific ELISpot responses of the T-Track CMV (Lophius Biosciences) and of the T-SPOT.CMV (Oxford Immunotec). This analysis included all pairs of data sets (parallel tests with T-Track CMV and T-SPOT.CMV) in liver transplant recipients (50 at the end of prophylaxis and 49–50 at month 1 thereafter), as also shown in Table 2. Panel (a) shows interferon-γ responses to CMV IE-1 antigens (n = 100) and panel (b) to CMV pp65 antigens (n = 99). At both time points, six out of 56 patients were not tested in parallel. The bold, continuous line indicates the regression line, the dashed lines the 95% confidence interval. SFU—spot forming units.

Figure 2

Comparison of CMV-specific ELISpot responses at the end of antiviral prophylaxis (white bars) and at month 1 thereafter (black bars). Results of the T-Track CMV (Lophius Biosciences, Lo) and of the T-SPOT.CMV (Oxford Immunotec, OI) in liver transplant recipients are shown as spot forming units (SFU) per 200,000 lymphocytes (T-Track CMV) and per 250,000 PBMC (T-SPOT.CMV) in panel (a–d). To allow a better comparison of the strength of reactions, results of the T-SPOT.CMV were further normalized to lymphocyte numbers and given as SFU per 200,000 lymphocytes, as shown in panel (e). Results were only considered if datasets for both ELISpot assays at the end of prophylaxis and at month 1 were available (47 out of 56 patients). Panel (a) and (e) show data on all liver transplant recipients (n = 47), panel (b–d) on patients with various combinations of CMV IgG in donors (D) and recipients (R). Please note the different scale on the y axis in panel (b). Mean and standard of the mean (SEM) are indicated. Data were compared by Wilcoxon matched pairs test (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 3

Course of responses in the QuantiFERON-CMV assay in 56 liver transplant recipients. Results of the QuantiFERON-CMV assay were analyzed from the end of antiviral prophylaxis until six months thereafter. The time after the end of prophylaxis correlated significantly with the concentration of IFN-γ, as analyzed by Spearman correlation analysis. The bold, continuous line indicates the regression line, the dashed lines the 95% confidence interval.

Figure 4

Discrimination between patients without and with CMV-DNAemia. This analysis considered CMV pp65-specific responses of the T-SPOT.CMV at the end of prophylaxis (a,b) and CMV IE-1-specific responses of the T-Track CMV at month 1 after the end of antiviral prophylaxis (c,d). For comparison, results of the QuantiFERON-CMV assay were shown (e,f). The cut-off for CMV-DNAemia was set at 500 copies/mL (substantial DNAemia). Panel (a,c,e) show results of receiver operating characteristic (ROC) curve analyses. It was analyzed if ELISpot or ELISA results were predictive of significant CMV-DNAemia. Panel (b,d,f) compare responses of the CMV pp65-ELISpot, CMV IE-1-ELISpot and QuantiFERON-CMV assay in patients with and without substantial CMV-DNAemia (Mann-Whitney U test). The horizontal lines indicate the mean values, the dashed line cut-off values as defined by ROC curve analyses (142 spot forming units (SFU), 3.5 SFU and 32 IU/mL, respectively). M = months after the end of prophylaxis.