Applicability of the cobas 6800 System for Epstein-Barr viral load quantitation using whole-blood specimens

Abstract

Objectives This study aimed to evaluate the analytical performance of the cobas 6800 System (Roche Diagnostics) and assess the feasibility of using whole-blood specimens instead of plasma.Methods The analytical performance of the cobas EBV test (Roche Diagnostics) was evaluated. Thereafter, 120 clinical samples were collected to compare the cobas EBV test and the artus EBV RG PCR Kit (Qiagen). The results of the cobas EBV test conducted using paired plasma as well as 5x and 10x diluted whole-blood specimens were compared with those of the artus EBV RG PCR Kit performed using whole blood.Results The precision of the cobas EBV test was acceptable, and its linearity was confirmed to be within the range of 2.85 to 6.89 log IU/mL. Cross-reactivity was not observed. The best qualitative agreement (Cohen kappa = 0.733) was observed using 5x diluted whole blood; the best quantitative correlation (Spearman correlation coefficient = 0.6865) was observed using 10x diluted whole blood.Conclusions A significant discrepancy was observed in the results obtained from the 2 assays because of the different specimens used. We observed, however, that diluting whole blood before conducting the cobas EBV test effectively resolved polymerase chain reaction inhibition and viscosity issues, leading to an acceptable correlation with the results from the artus EBV RG PCR Kit conducted using whole blood.