Background. Interferon-free treatment of chronic hepatitis C virus (HCV) genotype 1 infection may be shortened to 8 weeks in treatment-naive, noncirrhotic patients with baseline HCV RNA levels of <4 or <6 million (M) IU/mL based on post-hoc analyses of phase 3 trial data. The applicability of these viral load thresholds in clinical practice is unknown.
Methods. Pretreatment and on-treatment serum samples (n = 740) from patients with HCV genotype 1 infection were included for HCV RNA analysis with 2 widely used assays, Cobas AmpliPrep/CobasTaqMan (CAP/CTM) and Abbott RealTime HCV (ART) assays.
Results. HCV RNA levels were significantly higher with CAP/CTM than with ART (overall difference, +0.11 log10 IU/mL; P < .001). In treatment-naive, noncirrhotic patients, discordance rates around the clinical cutoffs at 4M and 6M IU/mL were 23% and 18%, respectively. The mean differences between assays in discordant samples were 0.38 (4M) and 0.41 (6M) log10 IU/mL, respectively. Overall, 87% and 95% of treatment-naive, noncirrhotic patients, respectively, had baseline HCV RNA levels below 4M and 6M IU/mL with ART. These rates were significantly higher than those measured with CAP/CTM (64% and 78%, respectively; P < .001). Finally, discordance rates around the proposed thresholds in 2 consecutive samples of the same patient were in the range of 1%–2% for ART and 13%–17% for CAP/CTM.
Conclusions. Selection of patients for 8-week regimens on the basis of a single HCV RNA determination may not be reliable because viral load levels around the proposed clinical thresholds show significant interassay and intrapatient variability.