Comparison of three real-time PCR for the quantification of polyomavirus BK.

Abstract Background Reactivation of latent polyomavirus BK is associated with nephropathy (PVAN) after renal transplantation. BK viral load determinations are a highly sensitive and specific method for predicting risk for PVAN. Objectives and study design The performance of three real-time PCR for BKV DNA quantification (MultiCode ® -RTx BK virus ASR [MC-RTx], MGB-Alert BKV ASR [MGB] and a laboratory developed assay [LDA]) were evaluated against a conventional PCR (test of record, TOR) in terms of linearity, dynamic range, and accuracy. Results The LOD (log 10 copies/ml) were 2.0, 2.0 and 3.0 for MC-RTx, MGB and LDA, respectively with a commercial plasma panel and 2.0, 2.6 and 3.5 with a urine panel. These assays demonstrated excellent linearity ( r 2  = 1.0) and reproducibility (CV range = 0.7–20.4%, 0.9–13.2%, and 0.5–13%, respectively). In an analysis of 100 clinical specimens, all 76 samples defined as true positive for BKV DNA (positive by two or more methods or a recent history of positivity) were detected with MC-RTx, while only 64 were detected with MGB and 55 were detected with LDA. BKV DNA was not detected by any method in the true negative specimens. Based on these results, the sensitivities were 100% for MC-RTx, 84% for MGB and 72% for LDA. The greatest linear correlation with the mean concentration was observed with MC-RTx ( r 2  = 0.96) with two samples (3%) with greater than 0.5 log 10 variance in quantification versus seven (11%) with MGB and ten (18%) with LDA. Conclusions These real-time assays for BKV load demonstrated excellent performance characteristics, with the MC-RTx demonstrating the greatest sensitivity.