A ccurate hepatitis C virus (HCV) RNA quantification is essential for the management and efficacy of treatment of chronic hepatitis C. The HCV RNA level is assessed using real-time PCRbased assays. Two highly sensitive commercial assays for HCV RNA quantification are available in many countries: the Roche Cobas AmpliPrep/Cobas TaqMan HCV assay (CAP/CTM HCV) (Roche Molecular Systems, Inc., Pleasanton, CA) and the Abbott RealTime HCV assay (ART HCV) (Abbott Molecular, Inc., Des Plaines, IL). Despite its good performance with most HCV strains, the CAP/CTM HCV test version 1.0 (v1.0) fails to detect genotype 4 strains with single nucleotide polymorphisms at positions 145 and 165 in the 5= untranslated region (5= UTR) (1). HCV genotype 4 is restricted to particular geographical areas, and many countries, including Japan, continue to use CAP/CTM HCV v1.0 to monitor HCV RNA quantification.We report two Japanese patients with HCV genotype 2a in whom HCV RNA was undetectable by CAP/CTM HCV v1.0, although hepatitis C viremia was confirmed by the ART HCV test (4.0 and 5.0 log 10 IU of HCV RNA/ml) and the Architect HCV core antigen assay (Abbott Diagnostics, Lake Forest, IL) (95 and 107 fmol/liter). This failure could be related to two or three substitutions in the putative binding site for the TaqMan probe (Fig. 1). The substitutions are at position 145, as described for HCV genotype 4 (1), and positions 158 and 169, which have not been reported previously.Underestimation of HCV genotype 2 RNA by CAP/CTM HCV v1.0 has been reported previously (2), but failure to detect HCV genotype 2a RNA is critical as this genotype is the second most common HCV genotype. Recently, a second version of the assay, CAP/CTM HCV v2.0 (3), with redesigned primers and an addi-