Evaluation of the fully automated LIAISON®XL chemiluminescence analyzer for QuantiFERON®-CMV testing in transplant recipients.

Abstract

Monitoring CMV-specific cell-mediated immunity by the QuantiFERON®-CMV (QF-CMV) may be useful in predicting the risk of CMV infection in transplant recipients (TR). As the QuantiFERON-Tuberculosis (QFT®-Plus) became available on the fully automated LIAISON®XL chemiluminescence (CLIA) analyzer, we evaluated the performance of the QF-CMV on the LIAISON®XL analyzer using the QuantiFERON®-TB Gold Plus reagent. Between 2018 and 2022, 81 samples from TR were collected at the Department of Virology of Limoges Hospital, France. Whole blood was collected into each of the three QF-CMV collection tubes: a CMV-antigen tube (QF-Ag), a mitogen tube (QF-Mg) (positive control), and a nil tube (negative control). The QF-CMV was performed on the LIAISON®XL analyzer, and results were compared with those obtained by conventional microplate ELISA. Intra- and inter-assay coefficients of variation were inferior to 20%. No inter-sample contamination was found (p=0.366). The level of concordance between CLIA and the commonly used ELISA method was 88.9%. Positive and negative percent agreements were 92.3% and 85.7%, respectively, with a very good agreement between assays (κ=0.818). Most discordances were due to indeterminate- or negative-ELISA/positive-CLIA results (most of ELISA results were borderline). Linear regression analyses demonstrated a strong correlation between both assays (QF-Ag Pearson's r=0.978, QF-Mg Pearson's r=0.963). No significant difference was observed in median QF-CMV values between both assays (QF-Ag p=0.776; QF-Mg p=0.853; Mann-Whitney U test). The Bland-Altman plots showed a minor difference in IFN-γ release (QF-Ag -0.069 IU/ml, 95% limits of agreement (LoA): -1.589; 1.451; QF-Mg0.190IU/ml, 95% LoA: -2.070; 2.450). Automated QF-CMV with CLIA is comparable to QF-CMV performed by ELISA with a presumably higher sensitivity for IFN-γ detection that may result in the conversion of samples close to the ELISA cut-off into positive results. Moreover, the use of a random-access analyzer allows to optimize the follow-up of TR.