Hepatitis C virus (HCV) infection is a major cause of chronic liver diseases, such as chronic hepatitis, cirrhosis and hepatocellular carcinoma.1 The use of molecular tests for HCV RNA detection has become very important in the management of infected patients.2 A variety of NAT for HCV RNA detection is now available in commercial kits, especially real-time quantitative PCR techniques based on TaqMan techniques.
In 1990s, in the absence of an international reference standard for HCV RNA, it was very difficult to compare results on HCV RNA levels from laboratories employing different commercial kits. The first World Health Organization (WHO) international standard for hepatitis C virus (HCV) RNA for nucleic acid amplification technology (NAT) (96/790) was established in 1997 based on the results of an international collaborative study, and its HCV RNA content was expressed in international units (IU).3 Since then, calibration of working reagents has become possible and harmonization of data from individual laboratories has been feasible. Subsequently, many laboratories adopted this standard to establish corresponding working standards for their daily work.4
An EQA scheme for the detection of HCV RNA was introduced by the National Center for Clinical Laboratories (NCCL) for HCV RNA in 1997. It comprises two proficiency panels annually, in which ten freeze-dried plasma specimens are distributed to participants with a request to report on the presence and concentration.
The commercial kits employed at that time were mainly for qualitative detection, so the results obtained by all the laboratories for HCV RNA detection from 1997 to 2002 were predominated by qualitative results.5 Thus, the National Center for Clinical laboratories began to develop a national reference material in 1998.6 The first national reference material of HCV RNA, developed referring to the WHO standard substance (NIBSC 96/790), had been approved to be the first-level standard substance of China in 2006. In the meantime, due to the application of a standard series of internal quality control and well-defined concentration reference materials to clinical laboratories and kits manufacturers, the precision of daytime detection in clinical laboratories could be effectively insured and the working standard of all manufacturers could be traced to the national reference.7 External quality assessment using lyophilized serum, in comparison with the WHO standard substance of HCV RNA, was applied in this study to discover problems so as to promote the improvement in the quality of HCV RNA detection.
Preparation of panels for quality control assessment
HCV positive plasma was diluted in plasma previously determined negative for hepatitis B virus (HBV), HCV and HIV. The positive plasma was diluted to the required concentration and divided in screw cap centrifuge tubes, 0.5 ml each to be lyophilized then stored at -40°C until distribution. Freeze-dried specimens were reconstituted with 0.5 ml molecular grade RNase free water prior to analysis. Each panel consisted of 5 coded samples. Three samples contained HCV RNA with different target levels. Two samples contained no virus and served as negative controls.
Composition and quantitative assays
Each panel consisted of 5 coded samples. Three samples contained HCV RNA detected by the automated method; COBAS Amplicor system (Roche) against WHO standard substance (NIBSC 96/790). The concentrations of the positive samples were set with multipoint comparison analysis on different concentrations of the WHO standard substance (NIBSC 96/790): the WHO standard substance and prepared positive samples underwent multipoint dilution and multiconcentration detection at the same time and then the multipoint detection data were processed and analyzed to derive the concentration. Assay procedures were as follows. (1) The WHO standard substance and prepared samples were first thawed with sterile RNase-free water. (2) The plasma negative for HBsAg, anti-HCV and anti-HIV were chosen to dilute the above-mentioned two standard substances, at dilution proportions of 1:3.2, 1:10, 1:32, 1:100 and 1: 320. (3) Specimens were extracted by the reagent specification, two prepared at each dilution. (4) The concentration of the positive samples was calculated according to the results of the WHO standard substance. (5) The WHO standard and analyte were diluted to the same dilutions and the content of internal the control was derived thereby from the concentration of the WHO standard; then the concentration of the corresponding analyte was calculated. A standard curve was plotted with the WHO standard substance (IU/ml) and the corresponding OD valuewho standard/OD valueinternal control to derive the concentration of the preparation or target concentration. Regarding the factors of extraction efficiency, amplification efficiency and probe combination etc. within the linear scope of this method, multiple points were chosen to be compared with the WHO standard.
The above mentioned panels were requested to be delivered to clinical laboratories and by express mail within the required time period. These laboratories were asked to apply their routine assay method for detection within the prescribed time period and deliver back the testing results on the report forms. Data were then put into the computer for statistical analysis. For evaluation of the quantitative data sets, the international units per milliliter results were first converted to log10 values. Then the overall geometric mean (GM) in log10 IU/ml and the standard deviation (SD) were calculated for each positive sample from all reported quantitative positive results. To assess the performance of individual participants, we calculated what percentage of the report positive results of each data set was within the acceptable range of target value (TG)±0.5 log10. This range was chosen because viral load differences of <0.5 log10 are usually not considered clinically relevant.
The number of laboratories reporting on qualitative results increased from 168 in 2003 to 233 in 2007. Most laboratories were reporting both qualitative and quantitative results with real-time PCR. The 5-year positive specimen results are shown in Table 1. The percentages of those laboratories totally correct from 2003 to 2006 were 79.6%, 79.7%, 75.5% and 74.9%, respectively and the percentage in 2007 was 84.2%.
For the positive samples, the negative result rates decreased with increasing sample viral load. As shown in Table 2, the total detection rate of specimens with the content ranging 5.25×103-6.61×103 IU/ml was 87.71%. However, the detection rate for specimens with the content above 106 U/ml was over 99%.
Commercial kits used by more than 5 laboratories were provided by the three manufacturers of Shanghai Kehua, Zhongshan Daan and Shenzhen PG (all in China). Specimens were statistically analyzed according to the different order of the content and the results showed that the undetected rate of reagents of the three manufacturers was related with the concentration (Table 3).
Since the samples for two panels, although codified differently, were the same, the interpanel reproducibility was assessed in overall results from each laboratory for each panel.
Specimens of two times of quality assessment in 2003 were all the same three positive specimens and the undetected rate is shown in Table 4.
Reported results of quantitative detection were rare in 2003 and the detection results by only a few laboratories could meet the required range (target value ±0.5 log) (Table 5, only 134 quantitative results for samples 0312, 0313 and 0315). With the increasing of the number of laboratories reporting on quantitative results in the following several years, results became more common until by the year of 2007, about 90% of all the laboratories met the required range for quantitative detection results. Summaries of the GM and SD for detection in each time are shown in Table 5. SD of low-concentration specimens was generally higher than that in the high-content specimens within the same year and that of similar-concentration specimens declined year by year.
It can be concluded from the above data that due to the effect of external quality assessment, results of all the laboratories improved yearly. As shown in Table 6, the national conformity rate statistically analyzed in 2003, when quantitative detection started, was only 30.5% and that in 2007 it had achieved 91.2%.
Detection results by major commercial kits
The close relationship between the results of laboratory detection and the employed kits and veracity and precision of the results for kit detection determined the success of laboratory detection. Table 7 shows the successive results for the two times of dilution in 2005 and 2007, the difference of (0.15–0.54) log between the GM and TG for primary commercial kits in 2005 and the difference of <0.3 log in 2007. In the meantime, SD for all reagents was decreased.
HCV RNA is one of the most efficient markers to confirm HCV infection and to assess the efficacy of the therapeutic regimens. This report describes the quality assurance program for the quantity and quantitative detection of HCV RNA in plasma for clinical laboratories in China. The application of clinical PCR detection in China may date back to the early 1990s, when commercial kits, laboratory settings, management and quality control were not appropriately standardized and false positive and false negative results frequently occurred. False positive results were generally correlated with amplicon contamination; false negative results, which related to the sensitivity of detection, were more often a problem. Almost half of laboratories could not get results from weak positive specimens with a content of 105 copies/ml;5 false negative conditions happened in domestic laboratories far more frequently than in overseas laboratories,9 which were closely related with the low sensitivity of reagents used in China. Confronting the problems in domestic clinical PCR detection, learning from foreign experience9 and combining that with the actual situation in China, our laboratories began to study how to conduct standardized management of clinical PCR laboratories in 1996. We established the external quality assessment system for HBV DNA and HCV RNA detection11,12 and we also investigated a variety of factors that affected clinical PCR detection.13,14 During 2000 and 2002, external quality assessment on qualitative detection of HCV RNA showed that the ratio of false positive declined from 26.7% in 2000 and 14.3% in 2001 to less than 8% in 2002.5
The reported lyophilized plasma for external quality assessment in this article was based on the research of the national standards for HCV RNA, the concentration of which was compared with the WHO standard substance.6 In 2003, the working standard was not compared with the national standards; they set the concentration on their own and copies/ml was taken as the unit. Our laboratory published the national standard with values fixed by comparing with the WHO standard substance (NIBSC 96/790) while conducting external quality assessment and provided a series of diluted lyophilized plasma7 for manufacturers to calibrate their working standards and clinical laboratories for them to use for internal quality control.
Since then a considerable effort has been put into developing more sensitive, reproducible and precise commercial PCR assays incorporating the IU as the standardize unit of measure. External quality assessment of qualitative detection of HCV RNA in 2003, which used the internationally accepted TG of ±0.5 log, derived poor results and only 30% of all the laboratories were within the required range. Nevertheless, the availability of the national standard for HCV RNA and the consequent adoption of the IU as a unit of measurement, allowed a more accurate assessment of the performance of the assay, with a better standardization of PCR method.
The quantitative assay showed more than 90% of laboratories' results were within the acceptable range in 2007. Moreover, the performance of the qualitative HCV RNA assay was not only enhanced for specimens with comparatively high content, but also for those with comparatively low content (3.00 log IU/ml). The accept rate (within the range of the TG±0.5 log) which was elevated to as much as 90% and SD being 0.68 indicate that laboratories got similar results even using different instruments, devices and kits and with different operators. The derived results were substantially improved compared with those in 2006 when the SDs were 1.51 and 1.16.
Rectifying working standards of commercial kits and ameliorating extraction methods of specimens at the same time, the accuracy and reproducibility of detection results were both improved. Especially through application of the kits of Kehua and PG the SD for detection of the content of specimens was below 0.3log in 2007, which was close to that of external quality assessment results of Italian scientists.15 Achievements in 2007 were significantly improved, mostly due to the use rate of 48% of these two reagent manufacturers kits.
Results of the external quality assessment indicate that in the past five years, with the implementation of standardized management in clinical PCR laboratories, the extensive application of a national standard and internal quality control plasma as well as the increasing improvement of kit quality, the quality of domestic clinical PCR detection of HCV RNA has been improving annually and false positive and false negative results have been remarkably decreased. On the other hand, some laboratories have had some difficulties in the exact quantification of the lowest and the highest viral levels (6.41 log10 IU/ml). It is evident that differences are greatest in specimens with lower concentrations (3.0 log10 IU/ml) of HCV and improvements in accuracy and standardization are required at the lower end of the test range.16 Moreover, PCR detection kits for HCV RNA should be further optimized by the manufacturers, in order to enhance the sensitivity and precision of detection. In addition, to ensure the effectiveness of clinical PCR detection of HCV RNA, clinical laboratories should further strengthen internal quality control, especially on consumables and kits for PCR detection, which will lay a good foundation for clinical PCR detection of HCV RNA.
In conclusion, our study indicates that considerable improvement of the molecular detection of HCV RNA has been achieved in recent years. However, further standardization is still needed. Finally, the present study underlines that external quality assessment is a very important tool for monitoring the quality of diagnostic laboratory tests.
1. Lauer GM, Walker BD. Hepatitis C infection. N Eng J Med 2001; 345: 41–52.
2. Majid AM, Gretch DR. Current and future hepatitis C virus diagnostic testing: problems advancements. Microbes Infect 2002; 4: 1227–1236.
3. Saldanha J, Lelie N, Heath A. Establishment of the first international standard for nucleic acid amplification technology (NAT) assay for HCV RNA. Vox Sang 1999; 76: 149–158.
4. Saldanha J, Heath A, Lelie N, Pisani G, Nubling M, Yu M. Calibration of HCV working reagents for NAT assay against the HCV international standard. The Collaborative Study Group. Vox Sang 2000; 78: 217–224.
5. Wang LN, Li JM, Deng W, Zheng HJ. External quality assessment
for detection of hepatitis C virus RNA by RT-PCR in clinical laboratories of China. Chin J Lab Med (Chin) 2003; 26: 777–779.
6. Wang LN, Wu JM, Li JM, Deng W, Wang ZF, Shen ZY, et al. Establishment of the first national standard for nucleic acid amplification technology (NAT) assay for HCV RNA. Chin J Lab Med (Chin) 2006; 29: 354–357.
7. Wang LN, Wu JM, Deng W, Shen ZY, Chen WX, Li JM, et al. Lyophilized standards for the calibration of real time PCR assay for hepatitis C virus RNA. Chin Med J 2006; 119: 1910–1914.
8. Schirm J, van Loon AM, Valentine-Thon E, Klapper PE, Reid J, Cleator GM. External quality assessment
program for qualitative and quantitative detection of hepatitis C virus RNA in diagnostic virology. J Clin Microbiol 2002; 40: 2973–80.
9. Gentili G, Pisani G, Bisso G, Cristiano K, Wirz M, Mele C. EQA Participants. Hepatitis C virus testing of plasma pools by nucleic acid amplification technology: external quality assessment
. Vox Sang 2001; 81: 143–147.
10. Wang LN, Deng W, Li JM. Two quantitative Hepatitis C Virus RNA reverse transcription PCR assays. Chin J Lab Med (Chin) 2001; 24: 146–147.
11. Wang LN, Li JM, Deng W, Zheng HJ. External quality assessment
for detection of hepatitis B virus DNA in clinical laboratories of China. Clin J Hepatol (Chin) 2004; 12: 501–502.
12. Wang LN, Li JM, Deng W, Ma R, Zheng HJ, Yang ZH. External quality assessment
of HCV RNA PCR detection in clinical laboratories of China. J Clin Lab Sci (Chin) 2000; 18: 281–283.
13. Wang LN, Zheng HJ, Deng W, Ma R, Li JM. Quality control of plasma (serum) specimen for the detection of HCV RNA with PCR. Chin J Hepatol (Chin) 1999; 7: 221–223.
14. Li JM, Wang LN, Deng W. Influence of nucleic acid extraction from serum with two methods on HBV DNA detection by fluorescence quantitative PCR (TaqMan). Chin J Lab Med (Chin) 2000; 23: 337–339.
15. Mancini C, Pisani G, Azzi A, Luisa Zerbini M, Gentili G, Mario Bisso G. MS participants. Inter-laboratory comparison of qualitative and quantitative detection of hepatitis C virus (HCV) RNA in diagnostic virology: a multicentre study (MS) in Italy. J Clin Virol 2004; 30: 313–319.
16. Chalker VJ, Rossouw A, Mee Z, Patel P, Vaughan H, James VL. External quality assessment
for the molecular detection of Hepatitis C virus. J Clin Virol 2007; 39: 141–144.