Background: The clinical utility of cytomegalovirus (CMV) viral load (VL) testing for managing transplant recipients and other immunocompromised patients has been established based on the performance of end-point PCR. However, due to the lack of well standardized tests, test results vary significantly across assays and laboratories. Here, we investigate key analytical performance characteristics of a new FDA-approved CMV VL test, while assessing differences at medically relevant VL measurements in patient samples.

Methods: Sensitivity and reproducibility of the COBAS® AmpliPrep/COBAS® TaqMan® CMV Test (TAQMAN) were determined using a panel derived from the 1st WHO International Standard. Linearity was determined in a dilution panel using AD169 cultured isolate and across CMV genotypes (Glycoprotein B, gt1-4). Specificity both in CMV-negative plasma and in the presence of potentially cross-reactive and interfering substances was assessed. Correlations to the COBAS® AMPLICOR CMV Test in 606 paired results were assessed using Deming Regression and Bland-Altman plots and the mean bias at different VL ranges (600-<2,000; 2,000-20,000; 20,000-<50,000; and >50,000 cp/mL) was calculated.

Results: The TAQMAN CMV test had a range of 137 – 9.1×10⁶ IU/mL across CMV glycoprotein B gt1-4 and was precise (S.D. <0.125 log₁₀ IU/mL between lots and <0.25 within all runs), accurate (nominal vs. observed range -0.04 to 0.09 log₁₀ across panel members spanning 91-1.8×10⁷ IU/mL) and sensitive (40.8 IU/mL, 95% PROBIT, C.I. 22.8-50.2) across genotypes tested. No interfering substances or cross reactivity with related species were found, and analytical specificity was 100%. Regression analysis showed good overall correlations (R² = 0.88 with parameter estimates intercept = -0.877; slope = 1.191) and Bland-Altman plots showing differences between paired results. Overall bias between methods at medically relevant result categories was within 0.2 log₁₀.

Conclusions: In this study, results generated from a new FDA approved real-time PCR test standardized the 1st WHO international standard showed good sensitivity and linearity across genotype 1-4, reproducibility and compared well to the AMPLICOR CMV Test (which was used to establish clinical utility of CMV DNA is several studies) in patient samples, suggesting the comparability of results for clinical decision making.

Cobb, B.: Employee, Roche Molecular Systems, Inc. Boisvert, D.: Employee, Roche Molecular Systems, Inc. Lee, S.: Employee, Roche Molecular Systems, Inc. Baum, P.: Employee, Roche Molecular Systems, Inc. Razonable, R.: Other, Roche Diagnostics Corporation, Advisory Board Honorarium. Caliendo, A.: Grant/Research Support, Roche Diagnostics Corporation. Yao, J.: Grant/Research Support, Roche Diagnostics Corporation. Do, T.: Employee, Roche Molecular Systems, Inc.

« Back to 2013 American Transplant Congress