HCV RNA measurement in samples with diverse genotypes using versions 1 and 2 of the Roche COBAS® AmpliPrep/COBAS® TaqMan® HCV test

Abstract

Background: Accurate HCV RNA measurement is required for monitoring treatment. Underquantification has been reported with some genotypes, particularly genotype 4, using version 1 of the Roche COBAS(®) AmpliPrep/COBAS(®) TaqMan(®) HCV Test.

Objectives: Compare the viral loads of clinical specimens representing diverse genotypes from across the United States using versions 1 (V1) and 2 (V2) of the Roche COBAS(®) AmpliPrep/COBAS(®) TaqMan(®) HCV Test. Assess the frequency of nt145 and nt165 variants in the 5' UTR associated with detection failures and underquantification in a large clinical sample database.

Study design: Three hundred archived clinical samples were measured using V1 and V2. Bland-Altman analysis was performed on log-transformed results and compared by genotype. The frequencies of nt145 and nt165 variants from 15,817 sequences were calculated.

Results: On average, V2 results were 0.16 logIU/mL lower than V1 results. The average genotype 4 sample was 0.08 logIU/mL higher in V1 than V2. The largest individual sample differences were -2.10 (genotype 2b) and 1.57 (genotype 2a) logIU/mL. For genotype 4 samples, the greatest underquantification by V1 was 1.46 logIU/mL. There were 13 (0.082%) variants at nt145 and 24 variants at nt165 (0.152%), including one sequence with variants at both positions (0.0063%).

Conclusions: Genotype 4 samples from the U.S. are rarely underquantified and not disproportionately so compared to other genotypes using the COBAS(®) AmpliPrep/COBAS(®) TaqMan(®) HCV Tests. Variants at the nt145 and nt165 positions are uncommon in the U.S. and double variants are exceedingly rare. Underquantification of HCV samples with the V1 assay is likely a very rare occurrence in U.S.

Keywords: Genotype; Hepatitis C virus; RNA; Viral load.