Home Search Abstracts Browse Sessions Program Committee View Session E-mail Abstract Author

 

 

Session 125 Poster Abstracts
Viral Load Assays
Thursday, 1:30 - 3:30 pm
Hall A


737    
Multi-laboratory Comparison of 2 Commercial Real-time HCV Viral Load Assays
Angela Caliendo*1, S Young2, A Gonzalez3, Y Zhou4, J Andersen4, A Valsamakis5, G Tsongalis6, B Yen-Lieberman7, R Pyles8, J Bremer9, and N Lurain9
1Emory Univ Sch of Med, Atlanta, GA, USA; 2Tricore Reference Labs, Albuquerque, NM, USA; 3Virginia Commonwealth Univ, Richmond, USA; 4Harvard Sch of Publ Hlth, Boston, MA, USA; 5Johns Hopkins Med Inst, Baltimore, MD, USA; 6Dartmouth-Hitchcock Med Ctr, Lebanon, NH, USA; 7Cleveland Clin Fndn, OH, USA; 8Univ of Texas Med Branch, Galveston, USA; and 9Rush Univ Med Ctr, Chicago, IL, USA

Background:  Viral load testing is the standard of care in monitoring response to hepatitis C virus (HCV) therapy in HCV/HIV co-infected individuals. Real-time PCR assays hold great promise for this application because they are very sensitive and have a broad linear range. However, the comparative performance of 2 commercial real-time viral load assays, the Abbott TaqMan ASR (Abbott) and the Roche TaqMan RUO (Roche) has not been evaluated. We report a multi-laboratory comparison on an evaluation of these two assays, to determine the limit of detection, linear range, reproducibility, and agreement.

Methods:  Plasma was obtained from an HCV-infected individual (genotype 1b) and the viral load was determined using the Roche Cobas Amplicor HCV Monitor test. This stock material was diluted in normal plasma to concentrations of 1 to 7 log10 IU/mL. Each of 4 panels had 7 replicates of each concentration, which was tested in 3 (Abbott) or 4 (Roche) labs. Data were log10 transformed prior to analysis.

Results:  Both assays detected all 28 replicates with a concentration of 1 log10 IU/mL and were linear to 7 log10 IU/mL. Overall the Roche assay was more reproducible than the Abbott assay, and both assays were less reproducible at lower concentrations of virus. For the Roche assay, the standard deviation (SD) was 0.18 and 0.09 log10 for replicates of 2 and 5.5 log10 IU/mL, respectively. While the SD for the Abbott assay was 0.42 and 0.26 log10 for replicates of 2 and 5.5 log10 IU/mL, respectively. For the Roche assay, the within-lab SD was below 0.15 log10 for all 4 labs at viral load viral values of ≥ 2.5 log10 IU/mL. For the Abbott assay, the within-lab SD ranged from 0.17 to 0.43 log10 at viral load values > 2.5 log10 IU/mL; for 3 of the 4 labs the SD was ≤ 0.15 log10 for viral load values ≥ 5.5 log10 IU/mL. In general, the viral load values obtained with the Roche assay were 0.2 to0.7 log10 greater than those obtained with the Abbott assay.

Conclusions:  The Roche assay can reliably detect 5-fold changes in viral load for values ≥ 2.5 log10 IU/mL, while this level of reliability is reached with the Abbott assay at values > 5.5 log10 IU/mL. Overall the sensitivity and linear range (at least 6 log10) of these real-time assays enables them to be used for both HCV diagnostics as well as therapeutic monitoring, which is not possible with the current RT-PCR assays.

Keywords: HCV; viral load; real-time assays