OBJECTIVE Viral serum concentrations are considered to have a clinical, prognostic and epidemiological impact on patients with hepatitis C infection. The purpose of this study was to test whether quantitation of HCV-RNA is possible by PCR in combination with DNA-ELISA. METHODS PCR with 25 to 35 cycles was performed with variable concentrations of cloned HCV-cDNA or the serum of patients with chronic hepatitis C. The amplified PCR-products were detected by agarose gel or by DNA-ELISA. RESULTS The detection limit of PCR with DNA-ELISA or gel detection decreased with increasing numbers of PCR cycles. However, the correlation of the optical density of the DNA-ELISA with the HCV-cDNA concentration decreased with increasing numbers of PCR as well (r=0.8 vs. r=0.29; 25 vs. 35 PCR-cycles). HCV-RNA was found in the sera of 19 of 30 patients (63%) with chronic hepatitis C by gel detection and in 14 of 30 patients (47%) by DNA-ELISA subsequent to PCR with 35 cycles. CONCLUSIONS The PCR/DNA-ELISA technique allows a semiquantitative determination of HCV-cDNA concentrations down to 103 genomes/ul. However, to obtain a reasonable sensitivity for HCV concentrations in the serum of patients with hepatitis C, the number of PCR cycles has to be increased to numbers too high to provide reliable quantification. Further studies should be done to evaluate whether the detection systems can be improved to obtain a sufficient sensitivity for quantitative HCV-PCR. A prerequisite for the use of PCR in combination with quantifiable detection systems is that a PCR-cycle number is chosen that keeps amplification within the logarithmic phase.