Evaluation of the Novel HISCL Chemiluminescence Enzyme Immunoassay for Laboratory Screening of Hepatitis C Virus

TEXT

About 2.8% of the world population has been infected with the hepatitis C virus (HCV), and it causes >350,000 deaths annually, but most people are unaware of their infections (1). Clinical laboratory tools play an essential role in the process of screening, diagnosis, and antiviral therapy assessment of HCV infection. These tools include serological testing for HCV antibodies or core antigen, nucleic acid testing of the HCV RNA level, and molecular testing of the HCV genotype (2). Of these, the anti-HCV antibody assay is the first-line screening test used to identify and engage HCV-infected persons (3). Recently, the HISCL Anti-HCV assay (Sysmex Corporation, Kobe, Japan) has been launched in China to provide services for clinical HCV screening. It employs chemiluminescence enzyme immunoassay (CLEIA) methodology, which uses recombinant proteins (NS3 and NS4, with amino acids 1186 to 1468 and 1688 to 1711, respectively) and synthetic polypeptides (core with amino acids 5 to 36 and 40 to 69) to detect specific HCV antibodies in human serum or plasma. Here, in order to verify its sensitivity and specificity, we appraised the performance of the HISCL Anti-HCV assay in a clinical laboratory in the West China region by using 10 commercial seroconversion panels and 1,048 routine clinical samples in parallel with the routinely used Elecsys Anti-HCV II assay (Roche Diagnostics, Penzberg, Germany).

This study was conducted with the approval of the Ethical Committee of West China Hospital of Sichuan University. The precision of the HISCL Anti-HCV assay was appraised according to CLSI document EP5-A2 in duplicate with two daily runs for 20 days (4). The precision within and between runs and total imprecision of the assay showed coefficients of variation (CVs) of 2.17, 3.12, and 5.16%, respectively, for quality control (QC) Viratrol level 1 (with a mean cutoff index [COI] of 6.55), CVs of 1.70, 2.16, and 5.05% for QC Viratrol level 2 (COI of 20.91), CVs of 1.41, 2.74, and 5.44% for weakly positive human pooled sera (COI of 10.00), and CVs of 1.29%, 1.90 and 5.11% for positive human pooled sera (COI of 47.85). In addition, when three sets of proficiency testing samples from the College of American Pathologists were used, the HISCL Anti-HCV assay demonstrated accurate results fully consistent with the official response.

The sensitivity of the assay for early anti-HCV detection was evaluated by using 10 different commercially available HCV seroconversion panels, including PHV912, PHV913, PHV914, PHV917 (M), HCV6212, HCV6224, HCV6227, HCV9045, HCV9058, and HCV10058, purchased from SeraCare Life Sciences Inc. (Milford, MA) and ZeptoMetrix Corporation (Buffalo, NY). The Elecsys Anti-HCV II assay detected 64.62% (42/65) of these seroconversion blood samples, while the HISCL Anti-HCV assay detected 46.15% (30/65). For four other anti-HCV assays whose results were given from the panel vendors, the rates of positive blood sample detection were 52.27% (23/44), 42.00% (21/50), 49.23% (32/65), and 50.77% (33/65) (Table 1). Furthermore, the HISCL Anti-HCV assay demonstrated a seroconversion sensitivity similar to that of HCV RNA when panels PHV913 and HCV10058 were used. Generally speaking, these finding suggest that the HISCL Anti-HCV assay had a seroconversion sensitivity for early HCV infection that was lower than that of the Elecsys Anti-HCV II assay but comparable to that of other widely used HCV screening assays.

In this study, 1,048 fresh serum samples from patients (mean age, 53 ± 16 years; 521 males and 527 females) at West China Hospital were tested with the Elecsys Anti-HCV II assay routinely and then with the HISCL Anti-HCV assay. Included were 700 fresh serum samples classified as anti-HCV antibody negative, 300 serum samples known to be anti-HCV antibody positive by the Elecsys Anti-HCV II assay, and 48 potentially interfering samples containing potential cross-reacting factors (5 from rheumatoid factor-positive patients, 5 from antinuclear antibody-positive patients, 3 from samples with hemolysis, 2 from patients with hyperlipidemia, 5 from patients with jaundice, 4 from patients with alcoholic cirrhosis, 5 from hepatitis B surface antigen-positive patients, and 19 from patients with other infections, such as HIV, Epstein-Barr virus, syphilis, and yeast infections).

According to other literature (5, 6), we designed the testing algorithms of serum samples shown in Fig. 1. Samples with discordant results by the two anti-HCV assays were investigated with the recomLine HCV IgG assay (Mikrogen Diagnostik, Neuried, Germany) and the Cobas AmpliPrep/Cobas TaqMan HCV test (Roche Diagnostics, Penzberg, Germany) for verification. The recomLine HCV IgG assay is a recombinant immunoblot assay that uses recombinant antigens from the core 1, core 2, helicase, NS3, NS4, and NS5 regions of HCV. For the HCV RNA assay, the lower limit of quantitation is 15 IU/ml. By using the testing algorithms, the result of a sample could be classified as negative, indeterminate, or positive. In this study, indeterminate results were excluded from further evaluation.

• Open in new tab
• Download powerpoint

FIG 1

Algorithm used to test clinical serum samples for HCV.

A total of 12 discordant samples were found. Three of these were confirmed as true positive, six were determined to be true negative, and three indeterminate samples were excluded from further analysis. Therefore, the overall concordance between the Elecsys Anti-HCV II assay and the HISCL Anti-HCV assay was 99.14% (1,036/1,045). Additionally, the kappa index was 0.9786 (95% confidence interval, 0.9647 to 0.9925), indicating that the agreement of the Anti-HCV assay results with those of the two assays was nearly perfect. As shown in Table 2, six false-positive results were obtained with the Elecsys Anti-HCV II assay and three false-negative results were obtained with the HISCL Anti-HCV assay. In 48 potentially interfering samples, no different qualitative result was shown with the two assays. In general, the HISCL Anti-HCV assay showed a satisfactory sensitivity of 98.97% and a specificity of 100%.

However, the HISCL Anti-HCV assay displayed a sensitivity comparable to that of other commercial anti-HCV assays, such as the InTec HCV EIA (with a sensitivity of 87.5%) (7), the Livzon Anti-HCV assay (87.5%) (7), the LIAISON XL Murex HCV Ab assay (93.18%) (8), the ADVIA Centaur HCV assay (97.5 to 100%) (9, 10, 11), the Architect Anti-HCV assay (94.44 to 100%) (5, 9, 10, 12), and the Ortho HCV 3.0 assay (98.3%) (6). This study suggests that the HISCL Anti-HCV assay could be optimized to reach superior sensitivity and to reduce the number of false-negative results, probably at the cost of lower specificity.

For clinical laboratories, other new features may make the HISCL Anti-HCV assay a unique choice for HCV screening, such as (i) a small sample volume requirement (20 μl), (ii) rapidity (17 min), (iii) high throughput (200 results/h), and (iv) continuous measurement. In conclusion, the HISCL Anti-HCV assay is a satisfactory assay with good precision, favorable sensitivity, and excellent specificity. It is suitable for routine use for the reliable detection of HCV antibodies and can serve as a novel choice for clinical HCV screening.