Switch from Signal Amplification to COBAS® AmpliPrep/COBAS® TaqMan 48: Is There a Need to Re-Baseline?

Madhuchhanda Choudhary; Gloria Caldito; Janice M Matthews-Greer

doi:10.29074/ascls.28.2.113

This article requires a subscription to view the full text. If you have a subscription you may use the login form below to view the article. Access to this article can also be purchased.

Madhuchhanda Choudhary, MD⇑
1. Department of Medicine, Division of Infectious Disease, Louisiana State University Health Sciences Center at Shreveport, Shreveport, LA
Gloria Caldito, PhD
1. Department of Biometry, Louisiana State University Health Sciences Center at Shreveport, Shreveport, LA
Janice M Matthews-Greer, PhD
1. Diagnostic Virology Laboratories, Louisiana State University Health Sciences Center at Shreveport, Shreveport, LA

Address for Correspondence: Madhuchhanda Choudhary. 1501 Kings Hwy, PO Box 33932. Shreveport, LA 71130. 318-675-5900. mchoud{at}lsuhsc.edu

Abstract

Background: Accurate quantitation of plasma Human immunodeficiency virus-1 (HIV) RNA levels is required for clinical management of HIV-1-infected patients. Several assays are used to quantify HIV RNA, and prior to implementing a change in viral load method, continuity in reported values is addressed by the laboratory and communicated to clinicians.

Methods: We initially compared COBAS® AmpliPrep-COBAS® TaqMan 48 (TaqMan) v.1.0 (Roche) to prior methodology, branched-chain DNA (bDNA) v.3.0 (Siemens) to determine if establishment of new patient baselines were necessary. Study data from 81 specimens run by both assays were compared using nonparametric tests, e.g., Wilcoxon signed-rank, Spearman. Subsets for comparison included only those that fell into overlapping ranges for both assays.

Results: The methods correlated (Spearman correlation, rs = 0.91), but TaqMan values were significantly higher than those of bDNA (p<0.0001). Based upon this, new baselines (n=768) collected over 6-months were determined by running bDNA on all specimens that could be quantitated by TaqMan (n=308). Of those with sufficient quantity to establish new baselines with the TaqMan, 308 and 272 were quantifiable by TaqMan and bDNA, respectively. Parallel data within overlapping ranges (n=262) were again highly correlated (rs =0.89), but still were statistically different (p=0.0044). Additional analyses for regression and pair differences by range were run on combined (study and parallel) data.

Conclusion: Our data demonstrate non-equivalence in HIV-1 RNA values of TaqMan v.1.0 as compared to bDNA v.3.0, and that new baselines for HIV viral load RNA HIV-1 should be re-established. New baselines for those patients missed by parallel testing can be calculated using regression analysis.