Performance evaluation of the DAAN HCV assay for quantification of hepatitis C virus RNA and its comparison with COBAS AmpliPrep/COBAS TaqMan HCV Quantitative Test, v2.0

Abstract Background The Daan HCV RNA quantitative assay was a recently developed kit with high sensitivity for the detection of HCV RNA. We aimed to evaluate the analytical performance of the Daan HCV RNA quantitative assay and compare it with the COBAS AmpliPrep/COBAS TaqMan HCV Quantitative Test, v2.0. Method WHO HCV RNA standard, NIBSC 06/102 standard, and CLSI EP documents were used to evaluate the precision, accuracy, linearity, anti‐interference ability, and cross‐reactivity of the Daan HCV RNA quantitative assay. Overall 198 clinical serum specimens were used to make comparison between the Daan HCV RNA quantitative assay and the Roche Cobas test. Results The within‐run precision (Swithin), and total precision (Stotal) for 6.11 log IU/mL, 4.22 log IU/mL, and 2.32 log IU/mL HCV RNA were 0.13 and 0.15, 0.07 and 0.09, and 0.11 and 0.10, respectively. The linear range was 20‐108 IU/mL, and the limit of detection was 15 IU/mL. It did not display any interference with commonly encountered conditions and cross‐reactivity with some common virus. A good agreement was observed between the Daan HCV RNA quantitative assay and the Roche Cobas test. Conclusion The Daan HCV RNA quantitative assay has shown satisfactory performances and excellent agreement with COBAS HCV Quantitative Test on clinical specimens with lower cost, which provides an alternative choice for the diagnosis and monitoring of HCV infection in developing countries.

cycle). The extraction of the HCV RNA was performed using automatic nucleic acids extraction apparatus Smart 32 (Daan), and the detection was carried out using the Applied Biosystems 7500 Realtime fluorescent quantitative PCR instrument (Thermo).

| Precision
To evaluate the precision, the World Health Organization (WHO) HCV RNA standard was used which contained 6.11 log IU/mL (1.3 × 10 6 IU/mL), 4.22 log IU/mL (1.66 × 10 4 IU/mL), and 2.32 log IU/mL (2.10 × 10 6 IU/mL), respectively. The American Clinical and Laboratory Standards Institute (CLSI) documents were used as guideline for the design of the experiments. According to CLSI document EP15-A2, 10 the three HCV RNA concentrations were tested in five replicates a day, respectively, and the detection lasted for 5 days. Within-run precision (S within ), variance term (B), and total precision (S total ) were calculated and compared with the within-run precision claimed by the manufacturer (σ within ) and total precision claimed by the manufacturer (σ total ). The computational formula for those parameters is shown in Supplementary Material.

| Accuracy
To evaluate the accuracy, the 6 log IU/mL HCV RNA reference (National Institute for Biological Standards and Control (NIBSC) 06/102 standard) was used and diluted with HCV-negative serum to generate five dilutions with nominal HCV RNA concentrations of 10 5 , 10 4 , 10 3 , 10 2 , 50 IU/mL and each dilution was tested in 3 replicates 10 .

| Limit of detection (LOD)
3 log IU/mL HCV RNA reference (NIBSC 06/102 standard) was diluted with HCV-negative serum to generate four dilutions with nominal HCV RNA concentrations of 50, 20, 15, 10 IU/mL, and each dilution was tested in 25 replicates. 12 The LOD was defined as the lowest HCV RNA level detected 95% of the times. 13

| Interference and cross-reactivity study
The interference experiment was carried out using the HCV RNA interfering substance kit (Daan), which took HCV RNA-positive serum | 3 of 7 HE Et al.
(WHO international standard, NIBSC code: 06/102) and five kinds of interfering substances as raw materials to generate serums with 30 mg/dL bilirubin, 3.2 g/dL triglyceride, 30 g/dL hemoglobin, 6 g/dL albumin, or 18 g/L total immunoglobulin G, containing 2.21 × 10 5 IU/ mL HCV RNA or 2.77 × 10 3 IU/mL HCV RNA. The two kinds of HCV RNA concentration serum with different interfering substances went through the HCV RNA quantitative assay to evaluate the effect of different interfering substances on the high concentration or low concentration HCV RNA-positive serum. 14 Moreover, the HCV RNA-negative serums with 30 mg/dL bilirubin, 3.2 g/dL triglyceride, 30 g/dL hemoglobin, 6 g/dL albumin, or 18 g/L total immunoglobulin G were also

| Precision and Accuracy
The S within , B, and S total of the three concentrations of HCV RNA were calculated and shown in Table 1. The manufacturer's claim with-run coefficient of variation (CV within ) and total coefficient of variation (CV total ) were both 5%. Thus, the σ within and σ total for 6.11 log IU/ mL, 4.22 log IU/mL, and 2.32 log IU/mL HCV RNA were 0.31, 0.21, and 0.12, respectively. Compared with the σ within and σ total , S within and S total of 6.11 log IU/mL and 4.22 log IU/mL were much less than the σ within and σ total , indicating excellent precision of high and medium concentrations of HCV RNA. However, the S within and S total of 2.32 log IU/mL were close to the σ within and σ total , suggesting the precision of low concentration HCV RNA was inferior to that of high and medium concentrations and variation was larger among assays with low concentration of HCV RNA. The SDs and CVs for different concentrations of HCV RNA were shown in Table 2. The CVs of 10 5 , 10 4 , 10 3 , and 10 2 IU/mL were all below 5%, while the CV of 50 IU/ mL was 11%. The CVs of each concentration indicated excellent accuracy. Moreover, the variation intended to become larger with the concentration decreasing.

| Linearity and limit of detection (LOD)
As shown in Figure 1, the Daan HCV RNA quantitative assay ex-  Table S1. The LOD was defined as the lowest HCV RNA level detected 95% of the times. As a result, the LOD was 15 IU/mL.

| Interference and cross-reactivity study
The results of the Daan HCV RNA quantitative assay for HCV RNApositive serum with different interfering substances were shown in Table 3  Bland-Altman analysis between the two assays ( Figure 2B), showing 77 quantitative results were within 95% limit of agreement among the 81 HCV RNA-positive specimens. The mean difference (bias ± SD) between the Daan HCV RNA quantitative assay and the Roche Cobas test was −0.07 ± 0.37 log IU/mL.

| D ISCUSS I ON
It is necessary for an HCV RNA quantitative assay to provide reliable results and wide range of quantification because the chronically infected patients had wide ranges of HCV viral load. 16 Moreover, broad-range quantification was needed to monitor the response of the HCV-infected patients to undetectable therapy levels of virus load. 17 The S within and S total of the Daan HCV RNA quantitative assay for each verified concentration had reached the standard claimed by the manufacturer, demonstrating excellent precision of the quantitative assay. The greatest variability was observed with the low (2.32 log IU/mL) HCV RNA concentration which may be caused by RNA degradation. According to the results of linearity, the Daan HCV RNA quantitative assay has a measurable range of at least eight orders of magnitude, generating adequate linearity up to 8 log IU/ mL (R 2 = 0.9976). The LOD of Daan HCV RNA quantitative assay was 15 log IU/mL which was comparable with COBAS AmpliPrep/ COBAS TaqMan HCV Quantitative Test whose LOD was also 15 log IU/mL. 18 Meanwhile, the HCV RNA quantitative assay with a 15 IU/mL limit of detection has been advocated by the Centers

for Disease Control of United States and European Society of Liver
Diseases as the confirmation test for HCV infection. 19 What's more, 700 μL serum or plasma was needed for COBAS AmpliPrep/COBAS TaqMan HCV Quantitative Test while only 200 μL serum or plasma was needed for the Daan HCV RNA quantitative assay, suggesting the Daan HCV RNA quantitative assay needed less specimen volume to get a comparable performance. What's more, the Daan HCV RNA quantitative assay was of high specificity for it did not display any interference with commonly encountered conditions and other viral illnesses. From all above, the DAAN quantitative showed good precision and accuracy and exhibited a wide range of quantification, a comparable LOD and high specificity, indicating its excellent analytical performance. Moreover, our study was the first to evaluate this recently developed and certified diagnostic kit for quantification of hepatitis C virus RNA.