ASSESSMENT OF PERFORMANCE OF QUALITY CONTROL SAMPLES USED IN REAL TIME QUANTITATIVE PCR FOR MONITORING VIRAL LOAD OF CMV, EBV, BKV IN ORGAN TRANSPLANT RECIPIENTS

Background: Precision in the measurement of CMV, EBV and BKV viral load are important for the management of solid organ transplant recipients. Using sample aliquots from a single batch with known copy numbers for quality control (QC) in all assay runs is an essential step to evaluate the precision and validate performance of each run. We report the QC data for the three real-time quantitative PCR (rt-QPCR) assays in our laboratory between Sept and Dec 2007

Methods: Laboratory-developed LightCycler rt-QPCR assays were used for monitoring viral load for CMV in plasma, EBV in whole blood, and BKV in plasma and urine, respectively. QC samples used in the three rt-QPCR assays are summarized below 

Results False positives results were identified in only 2 EBV runs. Inter-run variation was greatest at the lower end of the dynamic range and in positive QC samples that were extracted. Variation was greater in QC samples constructed using plasmid DNA than in QC samples constructed using virus or latently infected cells requiring extraction. Inter-run variation measured as CV (%) for the three QC samples in the CMV runs (130 runs) was 2.55, 4.89, and 2.83, respectively. 87% of the QCP1, 76%  of the QCP2 and 89%  of the QCP3 results fell within the acceptable range of variation relative to expected values. Inter-run variation measured as CV (%) for the three positive QC samples in the BK runs (79 runs) was 1.95, 1.80, and 4.45, respectively. 93.5% of the QCP1, 86% of the QCP2 and 67% of the QCP3 results fell within the acceptable range of variation relative to expected values. 88% to 99% of all results for the six positive control samples for the EBV runs (76 runs) fell within the acceptable range of variation relative to the expected values. 

Conclusions: Review of the QC performance in the CMV, EBV and BK QPCR assays allows us to control variation of quantitative assessments between runs and characterize individual runs as valid. However, 1 to 33% of QCs remain outside the acceptable range of variation. The study also implies that some variation is intrinsic to all rt-QPCR assays and it is important that clinicians interpret results with an understanding of this intrinsic variability.