s / Journal of Clinical Virology 69 (2015) 223–246 227 evaluated with serial dilutions (106–100 IU/mL; Roche). The quantitative correlation and lower limit of detection were evaluated by use of clinical samples. Findings: Picodroplet digital PCR showed a high degree of linearity (R2 =0.9891) and accuracy (SD≤0.31 log copies/mL; CV≤15%) across their detectable ranges. The quantitative correlation between the picodroplet digital PCR and the Roche TaqMan HCV test was high (R2 =0.9439) across their detectable ranges (106–15 IU/mL). We also found discordance in detection of low HCVRNA samples between the twomethods. In 18 clinical samples (HCV RNA 15 copies/mL, 44.4% with an HCV RNA levels < 15 copies/mL, and in 33.3% of samples the target was not detected. The discordance in low RNA samples needs to be verified using a large sample size. Interpretation: Picodroplet digital PCR could provide a useful method for HCV quantification, especially for low HCV RNA levels, but the approach needs to be further optimised. http://dx.doi.org/10.1016/j.jcv.2015.06.018