Transfusion Medicine
|
ORIGINAL ARTICLE
Improving the safety of blood transfusion by using a combination of two screening assays for hepatitis C virus K. Zhang, L. Wang, Y. Sun, R. Zhang, G. Lin, J. Xie & J. Li National Center for Clinical Laboratories, Beijing Hospital, Beijing, People’s Republic of China Received 22 April 2014; accepted for publication 19 August 2014
SUMMARY
Tel.: 86-10-58115061; fax: 86-10-65212064; e-mail: [email protected]
Hepatitis C virus (HCV) infection is one of most common infectious diseases globally, with 130 to 150 million people worldwide being chronic infected with HCV (World Health Organization, 2013). HCV infection leads to a high risk of developing liver cirrhosis and hepatocellular carcinoma, resulting in high mortality, morbidity and financial burden and is thus considered a global health problem (Kleinman et al., 2003; Nkrumah et al., 2011). HCV is transmitted primarily via blood and is recognised as the primary cause of transfusion-associated non-A–non-B viral hepatitis worldwide (Houghton et al., 1991). For these reasons, screening of donated blood for HCV is a routine practice (Nkrumah et al., 2011). The development of enzyme immunoassays (EIAs) with improved performance for the detection of anti-HCV allowed for rapid reduction of transfusion-transmitted (TT) HCV to 0·01% (per unit transfused) throughout the 1990s (Courouce et al., 1994); it is presently a recommended practice in blood banks around the world (Centers for Disease Control and Prevention, 1991). The adoption of nucleic acid amplification technology (NAT) at blood centres in the late 1990s shortened the window period necessary to detect anti-HCV antibodies (Stramer et al., 2004; Busch et al., 2005; Strobl, 2011; Selvarajah & Busch, 2012). A large number of countries worldwide have introduced NAT for HCV testing, which is performed in parallel to a single serological assay that allows discernment of the ‘stage’ of HCV-infected donors and safeguards the donated blood that centres receive annually (Department of Health and Human Services Food and Drug Administration Center for Biologics Evaluation and Research, 2004; Strobl, 2011; Selvarajah & Busch, 2012). Owing to the use of improved screening reagents, implementation of NAT, and strict donor selection procedures, the residual risk of TT-HCV declined over the last decade (Kim et al., 2012). The current risk of TT-HCV in the United States is down to one in a million (0·0001%) per unit transfused and is similarly low in most of the developed world (Selvarajah & Busch, 2012). However, the adequacy of screening donated blood by using a single serological assay in combination with NAT to sufficiently minimise the residual risk of TT-HCV is worth investigating. Although most individuals exposed to HCV develop chronic infection with detectable viremia, the infection resolves
© 2014 British Blood Transfusion Society
First published online 29 September 2014 doi: 10.1111/tme.12152
Background and Objectives: To illustrate that the combination of a single nucleic acid amplification test (NAT) with a single immunoassay for hepatitis C virus (HCV) detection, as proposed internationally, may lead to the omission of anti-HCV reactive sera with non-reactive NAT results. Materials and Methods: In total, 822 of 519, 299 serum samples from 11 blood centres in China were retested for anti-HCV by using 10 screening assays to detect HCV antibodies. A recombinant immunoblot assay (RIBA HCV 3·0; Ortho-Clinical Diagnostics) was performed to define confirmed HCV infection status. Samples with positive or indeterminate RIBA-HCV results were tested by quantitative tests for HCV RNA (Roche Diagnostics). Results: We found that 47 of the 822 (5·72%) serum samples were RIBA-positive without detectable HCV RNA. For these samples, the 10 anti-HCV immunoassays gave discordant and unsatisfactory results (detection rate ranging from 10·64 to 34·04%; ratio per 100 000 donations ranging from 5·97 to 8·09). Compared with a single anti-HCV screening assay, the two-assay combination increased the detection of these samples. The five best combinations [Sorin and Lizhu enzyme immunoassays (EIAs), Ortho and Lizhu EIAs, Sorin and Wantai EIAs, Sorin EIA and Roche CIA and Ortho and Wantai EIAs] increased the detection rate from 46·81 to 55·57%, thus reducing the ratio per 100 000 donations of HCV-seropositive samples. Conclusion: The combination of two anti-HCV screening immunoassays in parallel with an HCV NAT is a better strategy for HCV detection in blood centres to improve the safety of blood transfusion. Key words: anti-HCV, blood donors, combination of two assays, detection rate, NAT.
Correspondence: Jinming Li, National Center for Clinical Laboratories, Beijing Hospital, No.1, Dahua Road, Dongdan, Beijing 100730, People’s Republic of China.
298 K. Zhang et al. in others, who remain seropositive, though not viremic (Messick et al. 2001; Seeff et al. 2000); approximately 20% of these individuals were infected with the HCV but did not exhibit viremia (Busch et al., 2006). Therefore, stringent assessment, implementation and performance of anti-HCV EIA screening assays could have a major impact on safeguarding blood donation. However, anti-HCV immunoassays may lack sensitivity, and batch variation or manufacturers’ design changes may result in varying performance characteristics (Centers for Disease Control and Prevention, 1998; Contreras, 2006; Scheiblauer et al., 2006; Kamili et al., 2012). The objective of the present study was to verify that anti-HCV positive blood without detectable RNA may not be identified using a single anti-HCV screening immunoassay in parallel with an HCV NAT, and to investigate whether a combination of two anti-HCV immunoassays could increase the detection rate of these samples.
MATERIALS AND METHODS
Table 1. Anti-HCV screening assays used in the study Assay and method
Manufacturer
Anti-HCV, EIA
Ortho-Clinical Diagnostics, Raritan, NJ, USA Sorin Diagnosis, Italian Abbott Laboratories, Abbott Park, IL, USA Shanghai Kehua Bio-engineering Co., Ltd., Shanghai, China Wantai Biological Pharmacy, Beijing, China Jinhao Pharmaceutical Limited by Share Ltd., Beijing, China Intec Xiamen Science and Technology Co., Ltd., Xiamen, China Zhuhai Lizhu Reagents Limited by Share Ltd., Zhuhai, China Ortho-Clinical Diagnostics, Raritan, NJ, USA Roche Diagnostics, Branchburg, NJ, USA
Anti-HCV, EIA Murex anti-HCV, EIA Anti-HCV, EIA
Anti-HCV, EIA Anti-HCV, EIA
Anti-HCV, EIA
Anti-HCV, EIA
Donor samples Ortho VITROS anti-HCV assay, CIA
Samples of donated blood were collected for anti-HCV screening from 11 blood centres in China (Beijing, Shanghai, Zhejiang, Liaoning, Qingdao, Jiangsu, Shenzhen, Shandong, Kunming, Chengdu and Changzhou) between January 2010 and January 2011. These blood centres were part of a NAT pilot programme initiated by the Chinese Ministry of Health. These centres used one NAT for HCV RNA and two EIAs for anti-HCV antibodies across China. According to the programme, two EIA tests for HCV antibodies were performed. The blood units found to be repeatedly reactive by using two kits were discarded, and the samples that showed reactive anti-HCV results (S/CO ratio ≥1·0) from only one of the two EIA tests were shipped on dry ice from the blood centres to the National Center of Clinical Laboratories (NCCL) in Beijing.
Screening assays For this study, anti-HCV EIA kits from eight different manufactures and two automated CIA tests were used to screen for HCV antibodies (shown in Table 1). Testing and result interpretation were carried out according to the manufacturer’s instructions. Either of the two EIAs used by different blood centres (not shown) was also used in the NCCL screening step.
RIBA Third-generation recombinant immunoblot assay (RIBA HCV 3·0; Ortho-Clinical Diagnostics, Raritan, NJ, USA) was performed to define the sample status. RIBA detected C100 (NS4), C33c (NS3), C22p (core) and NS5. The assays were interpreted according to the manufacturer’s recommendations and deemed positive when two or more bands showed reactivity, indeterminate when only one band was reactive and negative for no reactivity.
Transfusion Medicine, 2014, 24, 297–304
Roche Elecsys
Eight anti-HCV EIA kits obtained from different manufactures and two automated CIA tests were used to screen HCV antibodies.
HCV NAT Quantitative HCV NAT was performed using the Roche COBAS AmpliPrep/COBASTaqMan HCV Test (Roche Diagnostics, Branchburg, NJ, USA) Testing and result interpretation followed the manufacturer’s instructions. The sensitivity (lowest limit of detection, LOD) of quantitative HCV NAT was 15 IU HCV RNA/mL.
Study design and statistical analysis The analytical steps of the HCV tests performed are illustrated in Fig. 1. The 822 samples that were shipped to NCCL were retested for HCV antibodies by using eight anti-HCV EIA kits and two CIA kits. All samples were also tested using a third-generation RIBA. Only those samples that showed positive or indeterminate RIBA-HCV results were subjected to quantitative HCV NAT. Confirmed HCV infection status was defined by reactive result in RIBA. Detection rates and incidence rate per 100 000 true HCV-seropositive samples as determined by one anti-HCV screening assay or two screening assays in combination were then defined.
RESULTS Results of donated blood sample testing A total of 519, 229 donations were tested for anti-HCV antibodies by using EIA during the study period at 11 blood centres.
© 2014 British Blood Transfusion Society
Combination of two screening assays for HCV
299
Tested for anti-HCV in blood centers n = 519,299 Reactive for anti-HCV according to one EIA kit and retested in NCCL by 10 assays n = 822 Tested by RIBA
Negative (n = 432)
HCV Ab-
Positive (n = 47)
HCV Ab+
Indeterminate (n = 343)
HCV Ab IND
Tested by HCV NAT
Reactive n=0
Non-reactive n = 390
Fig. 1. Schematic illustration of the analytical steps involved in HCV tests.
Of the samples screened with two EIA tests, 822 were found to be reactive by using only one of the two EIA kits (the reactivity rate was 0·158%) and were retested by using eight EIA and two CIA tests in NCCL, as described above. These samples were then subjected to RIBA, which yielded the following results: 432 samples (52·55%) were negative, 47 samples (5·72%) were positive and 343 samples (41·73%) were indeterminate. Of the positive and indeterminate samples, all samples subjected to HCV RNA testing were negative. The true reactive rate was 5·7% (47 of 822) by using RIBA as the gold standard (Fig. 1). RIBA patterns for 47 positive samples are shown in Table 2 and the total of RIBA patterns was eight. The sample numbers in four of them were fewer( 0·05).
samples might escape detection because the detection rates were relatively low (from 10·64 to 34·04%) for all eight EIA and both CIA kits. This indicates that using a single anti-HCV immunoassay screening test in parallel with a single NAT may result in the omission of true HCV-seropositive samples with non-reactive NAT results. There was no difference in the reactivity detection rates of RIBA-positive samples among the 10 assays (𝜒 2 test, P > 0·05). Table 3 also illustrates the incidence rates per 100 000 blood donations (5·97–8·09 per 100 000) for these true HCV-seropositive samples detected by 10 assays. The S/CO ratio distribution for eight EIAs showed no significant characteristics for eight RIBA patterns of 47 RIBA-positive samples. S/CO ratio distribution (>3·000) of any of the eight EIAs showed no significantly higher signal strength than S/CO ratio distribution (1·000–3·000) (Table 4).
Results of combination of two screening assays Results of a single screening assay Table 3 shows that among 47 confirmed anti-HCV positive samples without detectable RNA, most true HCV-seropositive
© 2014 British Blood Transfusion Society
After completing these preliminary steps, we proceeded to define detection rates of true HCV-seropositive samples by using two screening assays in combination. According to the results,
Transfusion Medicine, 2014, 24, 297–304
300 K. Zhang et al. Table 2. RIBA patterns of positive samples Number of detected samples for RIBA pattern of positive samples (%)
RIBA pattern C100/C33c/C22p (NS3/NS4/core) C100/C22p (NS4/core) C33c/C22p (NS3/core) C100/C33c (NS3/NS4) C33c/C22p/NS5 (NS3/core/NS5) C100/NS5 (NS4/NS5) C22p/NS5 (NS5/core) C33c/NS5b (NS3/NS5) Total
2 (4·26) 9 (19·15) 9 (19·15) 8 (17·02) 1 (2·13) 2 (4·26) 3 (6·38) 13 (27·66) 47
Eight RIBA patterns were obtained from 47 positive samples. Fewer samples ( 0·05).
Table 3. Results of 10 screening assays for RIBA-positive samples Assay1 A B C D E F G H I J Total
Number of positive samples detected
Detection rate (%)
Ratio per 100 000 donations
13 15 12 16 10 7 5 10 10 13 47
27·66 31·91 25·53 34·04 21·28 14·89 10·64 21·28 21·28 27·66
6·55 6·16 6·74 5·97 7·12 7·70 8·09 7·12 7·12 6·55
1 A,
B, C, D, E, F, G, H, I and J represent 10 assays for the detection of anti-HCV. A, Ortho-Clinical Diagnostics; B, Sorin Diagnostics; C, Wantai Biological Pharmacy; D, Zhuhai Lizhu Reagents Limited by Share Ltd.; F, Beijing Jinhao Pharmaceutical Limited by Share Ltd.; G, Intec Xiamen Science and Technology Co. Ltd; H, Shanghai Kehua Bio-engineering Co. Ltd.; I, Abbott Laboratories; E, Roche Elecsys; J, Ortho Vitros. E and J were automated anti-HCV CIAs, while the other eight assays were EIAs.
combination of any 2 of the 10 screening assays could increase detection rate for HCV RIBA-positive samples. Interestingly, it was found that the detection rates obtained using a combination of two assays was higher, as shown in Table 5. For example, Table 5 illustrates the best and least successful results of detection rates and the incidence rate for the two screening assays used in combination to detect true HCV-seropositive samples. The combination of Sorin and Lizhu EIAs, Ortho and Lizhu EIAs, Sorin and Wantai EIAs, Sorin EIA and Roche CIA and finally Ortho and Wantai EIAs were the five best combinations. Additionally,
Transfusion Medicine, 2014, 24, 297–304
the detection rates obtained using the two assays in combination (from 46·81 to 55·57%) and those obtained using either of the two assays used in the combination (𝜒 2 test, P < 0·05) were found to differ. This resulted in a reduction in the ratio per 100 000 HCV-seropositive samples, for example, combination of Sorin and Lizhu EIAs resulted in a 40·58 and 38·7% reduction (3·66 per 100 000) in the ratio per 100 000 donations compared to the use of Lizhu or Sorin individually (5·97 or 6·16 per 100 000). In contrast, the least successful combinations of two screening assays did not improve the overall detection ability (𝜒 2 test, P > 0·05); the detection rates ranged from 14·89 to 28·66%. In addition, three rounds of screening assays with NAT could not increase detection rates dramatically, for example, the best pattern obtained by using three screening assays was a combination of Ortho, Lizhu and Sorin EIAs. The number of positive results detected from the true HCV-seropositive samples was 30 (68·83%). No difference in detection rate was observed when a combination of Sorin and Lizhu EIAs (𝜒 2 test, P > 0·05) was used.
DISCUSSION The improvement of serological screening for HCV and the implementation of NAT technology prompted an investigation into the real status of HCV infection in blood donors. Our present study showed that anti-HCV RIBA-positive sera without detectable RNA might be missed when only one anti-HCV screening test (EIA or CIA) is used in parallel with NAT in Chinese blood donors. The combination of one NAT and two anti-HCV screening immunoassays provides a better strategy. NAT can identify HCV-infected donors early in the infectious window period and improve safe transfusion. However, serum HCV RNA levels fluctuate during chronic infection with intermittent viremia, yielding false-negative NAT results (Lanotte et al., 1998; Messick et al., 2001). In such cases, the viral load may be insufficient to elicit the full host response, resulting in a low level of anti-HCV antibodies. Another potential factor is that RIBA-positive and RNA non-reactive donors may be viremic below the NAT detection level, or may represent cases where HCV RNA exists intrahepatically and cannot be detected in circulation (Haydon et al., 1998; Messick et al., 2001; Lefrère et al., 2004). The virus spontaneously clears out of the body after infection in 15–20% individuals, who then remain negative for HCV RNA for a long time and show positive antibody test results in the absence of circulating virus. Although antibody reactivity declines over time after spontaneous resolution of infection, T-cell responses might be maintained. A previous study (Semmo et al., 2005) has shown that in the absence of HCV RNA, 50% RIBA-indeterminate blood donors had HCV-specific T-cell responses, as determined by sensitive ex vivo analyses of T-cell response, and these responses were typically focused on core peptides. In another study, previous T-cell response to HCV recombinant proteins (core, NS3 and NS3 helicase) (Bes et al., 2009) was analysed using an interferon (IFN) enzyme-linked immunospot assay and the cytokines in culture supernatants
© 2014 British Blood Transfusion Society
Combination of two screening assays for HCV
301
Table 4. S/CO ratio distribution for eight screening EIAs for RIBA-positive samples EIAs S/CO
A
B
C
D
F
G
H
I
3·000
34 7 6
32 10 5
35 8 4
31 12 4
40 5 2
42 1 4
37 7 3
37 8 2
Table 5. Results of screening assay combinations Combination type Best
Worst
1 A,
1 2 3 4 5 6 7 8 9 10
Primary assay1 Second assay1 Number of positive samples detected Detection rate (%) P-value2 Ratio per 100 000 donations B A B B A F E C H F
D D C E C H F G G G
28 26 23 22 22 13 13 13 12 7
55·57 55·32 48·94 46·91 46·81 28·66 27·66 27·66 25·53 14·89
0·05
3·66 4·04 4·62 4·81 4·81 6·55 6·55 6·55 6·74 7·70
B, C, D, E, F, G, H, I and J represent the same assays as those described for Table 1. in the detection rates of two-assay combinations with that either of individual assays put together (𝜒 2 test).
2 Differences
from donors with different RIBA results were quantified. The results confirmed that in the absence of HCV RNA, approximately half of RIBA-indeterminate donors have a resolved, previous HCV infection. These findings suggest that indeterminate donors should be counselled appropriately with respect to potential previous HCV exposure. Therefore, donations of HCV RNA-negative with discrepant anti-HCV results might transmit infection and could be significant for transfusion safety. In the present study, none of the 47 donations HCV-RIBA positive (5·7%, 47 of 822) was NAT reactive; these samples might represent the type of case discussed above. Since the development of third-generation assays, several commercial EIA or CIA kits for the detection of HCV antibodies in plasma have become available. These kits use recombinant proteins containing linear epitopes from both structural and non-structural regions of HCV proteins (Lin et al., 2005). However, these kits do not always perform equally well. This may be because of a large degree of variation in the individual’s immune response to the different antigens used in the kits or the different vectors used to clone the various recombinant proteins (Zachary et al., 2005). Although most anti-HCV immunoassay kits contain NS3, NS4, NS5 and core recombinant antigens, some of the recombinant antigens could not react with antibodies’ immunologic response to real donor samples. A previous study (Couroucé et al., 1995) has shown that no current HCV screening test is able to detect all infectious and antibody-positive specimens in spite of technological improvements. The previous results obtained from seven anti-HCV screening tests, including Ortho, Sorin and
© 2014 British Blood Transfusion Society
Murex, show different numbers of false-negative results. In the present study, of the 47 RIBA-positive sera without detectable RNA, 10 screening assays for anti-HCV also yielded discordant results (detection rates ranging from 10·64 to 34·04%) that were unsatisfactory, and most reactive samples went undetected by any one kit. Thus, the issue regarding undetected samples could not be resolved by using just one screening assay in parallel with an HCV NAT. The missed detection of these 47 samples of donated blood could have an impact on the safety of blood transfusion. The second anti-HCV screening test could offer an alternative for supplementing or confirming results, and the UK Health Protection Agency has published an algorithm that advised use of a second EIA for confirmation (National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C: 2002 June 10-12, 2002). In 2013, US CDC issued an updated guidance (Centers for Disease Control and Prevention (CDC), 2013) for laboratory testing and reporting of HCV antibodies owing to new developments, including the discontinuation of RIBA, which has been used for supplemental testing of HCV antibodies. The guidance recommended using a second HCV antibody assay different from the first, as a supplement test for HCV antibody. The strategy of using a combination of two screening assays for anti-HCV detection has been validated in a routine clinical laboratory (Allain et al., 1996) and evaluated in healthy blood donors in early 1996 (Evaluations and Standards Laboratory 2005). These results mainly focused on the confirmation of true reactive sera and focused on avoiding incorrect exclusion of healthy blood donors; however, reactive
Transfusion Medicine, 2014, 24, 297–304
302 K. Zhang et al. 2 screening test for anti-HCV
All reactive
HCV Ab+
Discardb
All negative
One negative One reactive
HCV Ab-
HCV Ab IND
Perform HCV NAT
Nonreactive
Report
Discard1, 2
Reactive
Discard
Fig. 2. Proposed algorithm for HCV testing in blood centres based on anti-HCV antibodies1 These donated blood samples should be removed from the blood supply or future donations from these donors should not be accepted, should be further confirmed.2 For these donors, additional testing as appropriate should be performed according to updated guidance from the CDC [Centers for Disease Control and Prevention (CDC), 2013].
sera could not be avoided, especially seropositive samples, without detectable RNA. It has been demonstrated (Zachary et al., 2005; Vermeersch et al., 2008) that using a confirmation assay with a different kit, developed by a different manufacturer will reduce the risk of both tests detecting the same false-reactivity. The present study showed that the combined usage of any two anti-HCV screening assays could improve the detection rate of RIBA-positive sera without detectable RNA. Using the best combination of two screening assays for anti-HCV detection, the detection rates of 47 samples were increased from 46·81 to 55·57% compared with that of individual assays (𝜒 2 test, P < 0·05). The incidence rates for these true HCV-seropositive samples detected by 10 assays (5·97 to 8·09 of 100 000) were reduced by using the best combination of Sorin and Lizhu EIAs (3·66 per 100 000). This showed that two screening assays in parallel with an HCV NAT could improve the safety of blood transfusion. In the present study, eight RIBA patterns were obtained for the 47 positive samples. The sample numbers in four of them were fewer ( 0·05). S/CO ratio distribution for eight EIAs showed no significant characteristics (Table 4). The relatively small number of positive samples was a limitation of this study. In addition, our study found that even the best three rounds of screening assays (Ortho, Lizhu and Sorin EIAs) with NAT could
Transfusion Medicine, 2014, 24, 297–304
not increase the detection rate significantly. No difference in detection rate was found between the best combination of three (68·83%) or two (55·57%) tests (𝜒 2 test, P > 0·05). Thus, there is no evidence suggesting that the safety of blood transfusion was increased by the combination of three or more anti-HCV kits; such combinations tend to increase the cost of screening. Thus, according to the results of this study, even with two screening assays in addition to NAT, the detection rate for the 47 samples tested still did not reach 100% (as shown in Table 2, the highest detection rate of the 47 confirmed positive samples was 55·57% by Sorin and Lizhu tests). That would translate into 19 RIBA confirmed reactive samples of 519 229 (3·66 of 100 000) still missed by two rounds of screening assays with NAT. It was shown here that there is no suitable combination to achieve 100% sensitivity rate to avoid viral transmission. Thus, undetected samples could not be avoided, and the sensitivity of commercial anti-HCV kits should be improved. Finally, our proposed algorithm for HCV testing in blood centres based on anti-HCV results is summarised in Fig. 2. Compared with algorithms proposed by previous reports (Chapko et al., 2005; Vermeersch et al., 2008; Lai et al., 2011) using single anti-HCV screening tests, our proposed screening for anti-HCV combines two screening assays. A similar sequential immunoassay strategy in conjunction with HCV NAT screening was study in Australian donors, where donors whose samples were reactive in the primary screening immunoassay were tested with a secondary immunoassay, and if found to be reactive using both assays, the samples were further tested by immunoblot (Kiely et al., 2004). The Chiron RIBA HCV 3·0 Strip Immunoblot Assay (Chiron Corp, Emeryville, CA, USA) that was recommended in
© 2014 British Blood Transfusion Society
Combination of two screening assays for HCV this study for supplemental testing of blood samples after initial HCV antibody testing is no longer available and the only other FDA-approved supplemental tests for HCV infection are those that detect HCV viremia (CDC, 2013). Therefore, in our proposed algorithm, if the results of both immunoassays are reactive, then the sample is regarded as anti-HCV positive and no further confirmation is necessary; the sample should be discarded in this case. Even if the result of one of the two assays is positive, the sample should be discarded. At this stage, it should be further confirmed whether these samples of donated blood should be removed from the blood supply and whether future donations from these donors should not be accepted as well. For these donors, additional testing as appropriate should be performed according to updated guidance by the CDC (CDC, 2013). On the other hand, if the results of both assays are negative, then the sample should be tested with NAT to assess the presence of HCV RNA. This approach would improve the detection rate of reactive samples and increase the safety of donated blood. In conclusion, based on our study, the use of a single NAT for HCV RNA and a single anti-HCV immunoassay may be
REFERENCES Allain, J.P., Kitchen, A., Aloysius, S., Petrik, J., Barbara, J.A. & Williamson, L.M. (1996) Safety and efficacy of hepatitis C virus antibody screening of blood donors with two sequential screening assays. Transfusion, 36, 401–405. Bes, M., Esteban, J.I., Casamitjana, N. et al. (2009) Hepatitis C virus (HCV)-specific T-cell responses among recombinant immunoblot assay-3-indeterminate blood donors: a confirmatory evidence of HCV exposure. Transfusion, 49, 296–305. Busch, M.P., Glynn, S.A., Stramer, S.L. et al. (2005) A new strategy for estimating risks of transfusion-transmitted viral infections based on rates of detection of recently infected donors. Transfusion, 45, 254–264. Busch, M.P., Glynn, S.A., Stramer, S.L. et al. (2006) Correlates of hepatitis C virus (HCV) RNA negativity among HCV-seropositive blood donors. Transfusion, 46, 469–475. Centers for Disease Control and Prevention (1991) Public Health Service inter-agency guidelines for screening donors of blood, plasma, organs, tissues, and semen for evidence of hepatitis B and hepatitis C. MMWR - Recommendations and Reports, 40, 1–17. Centers for Disease Control and Prevention (1998) Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCV-related chronic disease. MMWR, 47, 1–33.
© 2014 British Blood Transfusion Society
303
insufficient to detect hepatitis C antibodies in blood samples that are non-reactive in the NAT. The combination of one NAT and two anti-HCV screening immunoassays is therefore of considerable importance for safety of blood donation.
ACKNOWLEDGMENTS This work was supported by the Chinese Ministry of Health Special Fund for Public Welfare (201002005) and the National High Technology Research and Development Program (863 Program) (2011AA02A116). We gratefully acknowledge all listed blood centres for the provision of samples. K. Z. and L. W. performed the research, J. L. designed the research study, Y. S. and R. Z. contributed essential reagents and equipments, J. X. and G. L. analysed the data, and K. Z. wrote the paper.
CONFLICT OF INTEREST The authors declare that they have no conflicts of interest relevant to the manuscript submitted to Transfusion Med.
Centers for Disease Control and Prevention (CDC) (2013) Testing for HCV infection: an update of guidance for clinicians and laboratorians, 62, 362-365. Chapko, M.K., Sloan, K.L., Davison, J.W., Dufour, D.R., Bankson, D.D., Rigsby, M., Dominitz, J.A. & Hepatitis C Resource Center Program, Department of Veterans Affairs. (2005) Cost effectiveness of testing strategies for chronic hepatitis C. American Journal of Gastroenterology, 100, 607–615. Contreras, A.M. (2006) Hepatitis C antibody: true or false? New diagnosis strategies. Revista de Investigación Clínica, 58, 153–160. Courouce, A.M., Le Marrec, N., Girault, A., Ducamp, S. & Simon, N. (1994) Anti-hepatitis C virus (anti-HCV) seroconversion in patients undergoing hemodialysis: comparison of secondand third-generation anti-HCV assays. Transfusion, 34, 790–795. Couroucé, A.M., Barin, F., Botté, C., Lunel, F., Maisonneuve, P., Maniez, M. & Trepo, C. (1995) A comparative evaluation of the sensitivity of seven anti-hepatitis C virus screening tests. Vox Sanguinis, 69, 213–216. Department of Health and Human Services Food and Drug Administration Center for Biologics Evaluation and Research (2004) Use of Nucleic Acid Tests on Pooled and Individual Samples from Donors of Whole Blood and Blood Components (including Source Plasma and Source Leukocytes) to Adequately and Appropriately Reduce the
Risk of Transmission of HIV-1 and HCV. U.S. Food and Drug Administration. Haydon, G.H., Jarvis, L.M., Blair, C.S., Simmonds, P., Harrison, D.J., Simpson, K.J. & Hayes, P.C. (1998) Clinical significance of intrahepatitis C virus levels in patients with chronic HCV infection. Gut, 42, 570–575. Houghton, M., Weiner, A., Han, J., Kuo, G. & Choo, Q.L. (1991) Molecular biology of the hepatitis C viruses: implications for diagnosis, development and control of viral disease. Hepatology, 14, 381–388. Kamili, S., Drobeniuc, J., Araujo, A.C. & Hayden, T.M. (2012) Laboratory diagnostics for hepatitis C virus infection. Clinical Infectious Diseases, 55 (Suppl. 1), S43–S548. Kiely, P., Kay, D., Parker, S. & Piscitelli, L. (2004) The significance of third-generation HCV RIBA-indeterminate, RNA-negative results in voluntary blood donorsscreened with sequential third-generation immunoassays. Transfusion, 44, 349–358. Kim, M.J., Park, Q., Min, H.K. & Kim, H.O. (2012) Residual risk of transfusiontransmitted infection with human immunodeficiency virus, hepatitis C virus, and hepatitis B virus in Korea from 2000 through 2010. BMC Infectious Diseases, 12, 160. Kleinman, S.H., Kuhns, M.C., Todd, D.S., Glynn, S.A., McNamara, A., DiMarco, A. & Busch, M.P., Retrovirus Epidemiology Donor Study. (2003) Retrovirus Epidemiology Donor Study. Frequency of HBV
Transfusion Medicine, 2014, 24, 297–304
304 K. Zhang et al. DNA detection in US blood donors testing positive for the presence of anti-HBc: implications for transfusion transmission and donor screening. Transfusion, 43, 696–704. Lai, K.K., Jin, M., Yuan, S., Larson, M.F., Dominitz, J.A., & Bankson, D.D. (2011) Improved reflexive testing algorithm for hepatitis C infection using signal-to-cutoff ratios of a hepatitis C virus antibody assay. Clinical Chemistry, 57, 1050–1056. Lanotte, P., Dubois, F., Le Pogam, S. et al. (1998) The kinetics of antibodies against hepatitis C virus may predict viral clearance in exposed hemophiliacs. Journal of Infectious Diseases, 178, 556–559. Lefrère, J.J., Girot, R., Lefrère, F., Guillaume, N., Lerable, J., Le Marrec, N., Bouchardeau, F. & Laperche, S. (2004) Complete or partial seroreversion in immunocompetent individuals after self-limited HCV infection: consequences for transfusion. Transfusion, 44, 343–348. Lin, S., Arcangel, P., Medina-Selby, A. et al. (2005) Design of novel conformational and genotype-specific antigens for improving sensitivity of immunoassays for hepatitis C virus-specific antibodies. Journal of Clinical Microbiology, 43, 3917–3924. Messick, K., Sanders, J.C., Goedert, J.J. & Eyster, M.E. (2001) Hepatitis C viral clearance and antibody reactivity patterns in
Transfusion Medicine, 2014, 24, 297–304
persons with haemophilia and other cogenital bleeding disorders. Haemophilia, 7, 568–574. National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C: 2002 June 10-12 (2002) Hepatology, 36, S3–S20. URL http://www3.interscience.wiley.com/cgi-bin /fulltext/106597945/PDFSTART (Accessed 2014/09/05). Nkrumah, B., Owusu, M. & Averu, P. (2011) Hepatitis B and C viral infections among blood donors. A retrospective study from a rural community of Ghana. BMC Research Notes, 12, 4529. Scheiblauer, H., El-Nageh, M., Nick, S. et al. (2006) Evaluation of the performance of 44 assays used in countries with limited resources for the detection of antibodies to hepatitis C virus. Transfusion, 46, 708–718. Seeff, L.B., Miller, R.N., Rabkin, C.S. et al. (2000) 45-year follow-up of hepatitis C virus infection in healthy young adults. Annals of Internal Medicine, 132, 105–111. Selvarajah, S. & Busch, M.P. (2012) Transfusion transmission of HCV, a long but successful road map to safety. Antiviral Therapy, 17 (7 Pt B),, 1423–1429. Semmo, N., Barnes, E., Taylor, C., Kurtz, J., Harcourt, G., Smith, N. & Klenerman, P.
(2005) T-cell responses and previous exposure to hepatitis C virus in indeterminate blood donors. Lancet, 365, 327–329. Stramer, S.L., Glynn, S.A., Kleinman, S.H. et al. (2004) Detection of HIV-1 and HCV infections among antibody-negative blood donors by nucleic acid amplification testing. New England Journal of Medicine, 351, 760–768. Strobl, F. (2011) NAT in blood screening around the world. MLO: Medical Laboratory Observer, 43, 12–14, 16, 18; quiz 20–21. Vermeersch, P., Van Ranst, M. & Lagrou, K. (2008) Validation of a strategy for HCV antibody testing with two enzyme immunoassays in a routine clinical laboratory. Journal of Clinical Virology, 42, 394–398. World Health Organization 2013. Hepatitis C Fact Sheet No. 164. (updated April 2014). URL http://www.who.int/mediacentre/ factsheets/fs164/en/ (Accessed 2014/09/05). Zachary, P., Ullmann, M., Djeddi, S., Meyer, N., Wendling, M.J., Schvoerer, E., Stoll-Keller, F. & Gut, J.P. (2005) Evaluation of three commercially available hepatitis C virus antibody detection assays under the conditions of a clinical virology laboratory. Journal of Clinical Virology, 34, 207–210; discussion 216-218.
© 2014 British Blood Transfusion Society
|
ORIGINAL ARTICLE
Improving the safety of blood transfusion by using a combination of two screening assays for hepatitis C virus K. Zhang, L. Wang, Y. Sun, R. Zhang, G. Lin, J. Xie & J. Li National Center for Clinical Laboratories, Beijing Hospital, Beijing, People’s Republic of China Received 22 April 2014; accepted for publication 19 August 2014
SUMMARY
Tel.: 86-10-58115061; fax: 86-10-65212064; e-mail: [email protected]
Hepatitis C virus (HCV) infection is one of most common infectious diseases globally, with 130 to 150 million people worldwide being chronic infected with HCV (World Health Organization, 2013). HCV infection leads to a high risk of developing liver cirrhosis and hepatocellular carcinoma, resulting in high mortality, morbidity and financial burden and is thus considered a global health problem (Kleinman et al., 2003; Nkrumah et al., 2011). HCV is transmitted primarily via blood and is recognised as the primary cause of transfusion-associated non-A–non-B viral hepatitis worldwide (Houghton et al., 1991). For these reasons, screening of donated blood for HCV is a routine practice (Nkrumah et al., 2011). The development of enzyme immunoassays (EIAs) with improved performance for the detection of anti-HCV allowed for rapid reduction of transfusion-transmitted (TT) HCV to 0·01% (per unit transfused) throughout the 1990s (Courouce et al., 1994); it is presently a recommended practice in blood banks around the world (Centers for Disease Control and Prevention, 1991). The adoption of nucleic acid amplification technology (NAT) at blood centres in the late 1990s shortened the window period necessary to detect anti-HCV antibodies (Stramer et al., 2004; Busch et al., 2005; Strobl, 2011; Selvarajah & Busch, 2012). A large number of countries worldwide have introduced NAT for HCV testing, which is performed in parallel to a single serological assay that allows discernment of the ‘stage’ of HCV-infected donors and safeguards the donated blood that centres receive annually (Department of Health and Human Services Food and Drug Administration Center for Biologics Evaluation and Research, 2004; Strobl, 2011; Selvarajah & Busch, 2012). Owing to the use of improved screening reagents, implementation of NAT, and strict donor selection procedures, the residual risk of TT-HCV declined over the last decade (Kim et al., 2012). The current risk of TT-HCV in the United States is down to one in a million (0·0001%) per unit transfused and is similarly low in most of the developed world (Selvarajah & Busch, 2012). However, the adequacy of screening donated blood by using a single serological assay in combination with NAT to sufficiently minimise the residual risk of TT-HCV is worth investigating. Although most individuals exposed to HCV develop chronic infection with detectable viremia, the infection resolves
© 2014 British Blood Transfusion Society
First published online 29 September 2014 doi: 10.1111/tme.12152
Background and Objectives: To illustrate that the combination of a single nucleic acid amplification test (NAT) with a single immunoassay for hepatitis C virus (HCV) detection, as proposed internationally, may lead to the omission of anti-HCV reactive sera with non-reactive NAT results. Materials and Methods: In total, 822 of 519, 299 serum samples from 11 blood centres in China were retested for anti-HCV by using 10 screening assays to detect HCV antibodies. A recombinant immunoblot assay (RIBA HCV 3·0; Ortho-Clinical Diagnostics) was performed to define confirmed HCV infection status. Samples with positive or indeterminate RIBA-HCV results were tested by quantitative tests for HCV RNA (Roche Diagnostics). Results: We found that 47 of the 822 (5·72%) serum samples were RIBA-positive without detectable HCV RNA. For these samples, the 10 anti-HCV immunoassays gave discordant and unsatisfactory results (detection rate ranging from 10·64 to 34·04%; ratio per 100 000 donations ranging from 5·97 to 8·09). Compared with a single anti-HCV screening assay, the two-assay combination increased the detection of these samples. The five best combinations [Sorin and Lizhu enzyme immunoassays (EIAs), Ortho and Lizhu EIAs, Sorin and Wantai EIAs, Sorin EIA and Roche CIA and Ortho and Wantai EIAs] increased the detection rate from 46·81 to 55·57%, thus reducing the ratio per 100 000 donations of HCV-seropositive samples. Conclusion: The combination of two anti-HCV screening immunoassays in parallel with an HCV NAT is a better strategy for HCV detection in blood centres to improve the safety of blood transfusion. Key words: anti-HCV, blood donors, combination of two assays, detection rate, NAT.
Correspondence: Jinming Li, National Center for Clinical Laboratories, Beijing Hospital, No.1, Dahua Road, Dongdan, Beijing 100730, People’s Republic of China.
298 K. Zhang et al. in others, who remain seropositive, though not viremic (Messick et al. 2001; Seeff et al. 2000); approximately 20% of these individuals were infected with the HCV but did not exhibit viremia (Busch et al., 2006). Therefore, stringent assessment, implementation and performance of anti-HCV EIA screening assays could have a major impact on safeguarding blood donation. However, anti-HCV immunoassays may lack sensitivity, and batch variation or manufacturers’ design changes may result in varying performance characteristics (Centers for Disease Control and Prevention, 1998; Contreras, 2006; Scheiblauer et al., 2006; Kamili et al., 2012). The objective of the present study was to verify that anti-HCV positive blood without detectable RNA may not be identified using a single anti-HCV screening immunoassay in parallel with an HCV NAT, and to investigate whether a combination of two anti-HCV immunoassays could increase the detection rate of these samples.
MATERIALS AND METHODS
Table 1. Anti-HCV screening assays used in the study Assay and method
Manufacturer
Anti-HCV, EIA
Ortho-Clinical Diagnostics, Raritan, NJ, USA Sorin Diagnosis, Italian Abbott Laboratories, Abbott Park, IL, USA Shanghai Kehua Bio-engineering Co., Ltd., Shanghai, China Wantai Biological Pharmacy, Beijing, China Jinhao Pharmaceutical Limited by Share Ltd., Beijing, China Intec Xiamen Science and Technology Co., Ltd., Xiamen, China Zhuhai Lizhu Reagents Limited by Share Ltd., Zhuhai, China Ortho-Clinical Diagnostics, Raritan, NJ, USA Roche Diagnostics, Branchburg, NJ, USA
Anti-HCV, EIA Murex anti-HCV, EIA Anti-HCV, EIA
Anti-HCV, EIA Anti-HCV, EIA
Anti-HCV, EIA
Anti-HCV, EIA
Donor samples Ortho VITROS anti-HCV assay, CIA
Samples of donated blood were collected for anti-HCV screening from 11 blood centres in China (Beijing, Shanghai, Zhejiang, Liaoning, Qingdao, Jiangsu, Shenzhen, Shandong, Kunming, Chengdu and Changzhou) between January 2010 and January 2011. These blood centres were part of a NAT pilot programme initiated by the Chinese Ministry of Health. These centres used one NAT for HCV RNA and two EIAs for anti-HCV antibodies across China. According to the programme, two EIA tests for HCV antibodies were performed. The blood units found to be repeatedly reactive by using two kits were discarded, and the samples that showed reactive anti-HCV results (S/CO ratio ≥1·0) from only one of the two EIA tests were shipped on dry ice from the blood centres to the National Center of Clinical Laboratories (NCCL) in Beijing.
Screening assays For this study, anti-HCV EIA kits from eight different manufactures and two automated CIA tests were used to screen for HCV antibodies (shown in Table 1). Testing and result interpretation were carried out according to the manufacturer’s instructions. Either of the two EIAs used by different blood centres (not shown) was also used in the NCCL screening step.
RIBA Third-generation recombinant immunoblot assay (RIBA HCV 3·0; Ortho-Clinical Diagnostics, Raritan, NJ, USA) was performed to define the sample status. RIBA detected C100 (NS4), C33c (NS3), C22p (core) and NS5. The assays were interpreted according to the manufacturer’s recommendations and deemed positive when two or more bands showed reactivity, indeterminate when only one band was reactive and negative for no reactivity.
Transfusion Medicine, 2014, 24, 297–304
Roche Elecsys
Eight anti-HCV EIA kits obtained from different manufactures and two automated CIA tests were used to screen HCV antibodies.
HCV NAT Quantitative HCV NAT was performed using the Roche COBAS AmpliPrep/COBASTaqMan HCV Test (Roche Diagnostics, Branchburg, NJ, USA) Testing and result interpretation followed the manufacturer’s instructions. The sensitivity (lowest limit of detection, LOD) of quantitative HCV NAT was 15 IU HCV RNA/mL.
Study design and statistical analysis The analytical steps of the HCV tests performed are illustrated in Fig. 1. The 822 samples that were shipped to NCCL were retested for HCV antibodies by using eight anti-HCV EIA kits and two CIA kits. All samples were also tested using a third-generation RIBA. Only those samples that showed positive or indeterminate RIBA-HCV results were subjected to quantitative HCV NAT. Confirmed HCV infection status was defined by reactive result in RIBA. Detection rates and incidence rate per 100 000 true HCV-seropositive samples as determined by one anti-HCV screening assay or two screening assays in combination were then defined.
RESULTS Results of donated blood sample testing A total of 519, 229 donations were tested for anti-HCV antibodies by using EIA during the study period at 11 blood centres.
© 2014 British Blood Transfusion Society
Combination of two screening assays for HCV
299
Tested for anti-HCV in blood centers n = 519,299 Reactive for anti-HCV according to one EIA kit and retested in NCCL by 10 assays n = 822 Tested by RIBA
Negative (n = 432)
HCV Ab-
Positive (n = 47)
HCV Ab+
Indeterminate (n = 343)
HCV Ab IND
Tested by HCV NAT
Reactive n=0
Non-reactive n = 390
Fig. 1. Schematic illustration of the analytical steps involved in HCV tests.
Of the samples screened with two EIA tests, 822 were found to be reactive by using only one of the two EIA kits (the reactivity rate was 0·158%) and were retested by using eight EIA and two CIA tests in NCCL, as described above. These samples were then subjected to RIBA, which yielded the following results: 432 samples (52·55%) were negative, 47 samples (5·72%) were positive and 343 samples (41·73%) were indeterminate. Of the positive and indeterminate samples, all samples subjected to HCV RNA testing were negative. The true reactive rate was 5·7% (47 of 822) by using RIBA as the gold standard (Fig. 1). RIBA patterns for 47 positive samples are shown in Table 2 and the total of RIBA patterns was eight. The sample numbers in four of them were fewer( 0·05).
samples might escape detection because the detection rates were relatively low (from 10·64 to 34·04%) for all eight EIA and both CIA kits. This indicates that using a single anti-HCV immunoassay screening test in parallel with a single NAT may result in the omission of true HCV-seropositive samples with non-reactive NAT results. There was no difference in the reactivity detection rates of RIBA-positive samples among the 10 assays (𝜒 2 test, P > 0·05). Table 3 also illustrates the incidence rates per 100 000 blood donations (5·97–8·09 per 100 000) for these true HCV-seropositive samples detected by 10 assays. The S/CO ratio distribution for eight EIAs showed no significant characteristics for eight RIBA patterns of 47 RIBA-positive samples. S/CO ratio distribution (>3·000) of any of the eight EIAs showed no significantly higher signal strength than S/CO ratio distribution (1·000–3·000) (Table 4).
Results of combination of two screening assays Results of a single screening assay Table 3 shows that among 47 confirmed anti-HCV positive samples without detectable RNA, most true HCV-seropositive
© 2014 British Blood Transfusion Society
After completing these preliminary steps, we proceeded to define detection rates of true HCV-seropositive samples by using two screening assays in combination. According to the results,
Transfusion Medicine, 2014, 24, 297–304
300 K. Zhang et al. Table 2. RIBA patterns of positive samples Number of detected samples for RIBA pattern of positive samples (%)
RIBA pattern C100/C33c/C22p (NS3/NS4/core) C100/C22p (NS4/core) C33c/C22p (NS3/core) C100/C33c (NS3/NS4) C33c/C22p/NS5 (NS3/core/NS5) C100/NS5 (NS4/NS5) C22p/NS5 (NS5/core) C33c/NS5b (NS3/NS5) Total
2 (4·26) 9 (19·15) 9 (19·15) 8 (17·02) 1 (2·13) 2 (4·26) 3 (6·38) 13 (27·66) 47
Eight RIBA patterns were obtained from 47 positive samples. Fewer samples ( 0·05).
Table 3. Results of 10 screening assays for RIBA-positive samples Assay1 A B C D E F G H I J Total
Number of positive samples detected
Detection rate (%)
Ratio per 100 000 donations
13 15 12 16 10 7 5 10 10 13 47
27·66 31·91 25·53 34·04 21·28 14·89 10·64 21·28 21·28 27·66
6·55 6·16 6·74 5·97 7·12 7·70 8·09 7·12 7·12 6·55
1 A,
B, C, D, E, F, G, H, I and J represent 10 assays for the detection of anti-HCV. A, Ortho-Clinical Diagnostics; B, Sorin Diagnostics; C, Wantai Biological Pharmacy; D, Zhuhai Lizhu Reagents Limited by Share Ltd.; F, Beijing Jinhao Pharmaceutical Limited by Share Ltd.; G, Intec Xiamen Science and Technology Co. Ltd; H, Shanghai Kehua Bio-engineering Co. Ltd.; I, Abbott Laboratories; E, Roche Elecsys; J, Ortho Vitros. E and J were automated anti-HCV CIAs, while the other eight assays were EIAs.
combination of any 2 of the 10 screening assays could increase detection rate for HCV RIBA-positive samples. Interestingly, it was found that the detection rates obtained using a combination of two assays was higher, as shown in Table 5. For example, Table 5 illustrates the best and least successful results of detection rates and the incidence rate for the two screening assays used in combination to detect true HCV-seropositive samples. The combination of Sorin and Lizhu EIAs, Ortho and Lizhu EIAs, Sorin and Wantai EIAs, Sorin EIA and Roche CIA and finally Ortho and Wantai EIAs were the five best combinations. Additionally,
Transfusion Medicine, 2014, 24, 297–304
the detection rates obtained using the two assays in combination (from 46·81 to 55·57%) and those obtained using either of the two assays used in the combination (𝜒 2 test, P < 0·05) were found to differ. This resulted in a reduction in the ratio per 100 000 HCV-seropositive samples, for example, combination of Sorin and Lizhu EIAs resulted in a 40·58 and 38·7% reduction (3·66 per 100 000) in the ratio per 100 000 donations compared to the use of Lizhu or Sorin individually (5·97 or 6·16 per 100 000). In contrast, the least successful combinations of two screening assays did not improve the overall detection ability (𝜒 2 test, P > 0·05); the detection rates ranged from 14·89 to 28·66%. In addition, three rounds of screening assays with NAT could not increase detection rates dramatically, for example, the best pattern obtained by using three screening assays was a combination of Ortho, Lizhu and Sorin EIAs. The number of positive results detected from the true HCV-seropositive samples was 30 (68·83%). No difference in detection rate was observed when a combination of Sorin and Lizhu EIAs (𝜒 2 test, P > 0·05) was used.
DISCUSSION The improvement of serological screening for HCV and the implementation of NAT technology prompted an investigation into the real status of HCV infection in blood donors. Our present study showed that anti-HCV RIBA-positive sera without detectable RNA might be missed when only one anti-HCV screening test (EIA or CIA) is used in parallel with NAT in Chinese blood donors. The combination of one NAT and two anti-HCV screening immunoassays provides a better strategy. NAT can identify HCV-infected donors early in the infectious window period and improve safe transfusion. However, serum HCV RNA levels fluctuate during chronic infection with intermittent viremia, yielding false-negative NAT results (Lanotte et al., 1998; Messick et al., 2001). In such cases, the viral load may be insufficient to elicit the full host response, resulting in a low level of anti-HCV antibodies. Another potential factor is that RIBA-positive and RNA non-reactive donors may be viremic below the NAT detection level, or may represent cases where HCV RNA exists intrahepatically and cannot be detected in circulation (Haydon et al., 1998; Messick et al., 2001; Lefrère et al., 2004). The virus spontaneously clears out of the body after infection in 15–20% individuals, who then remain negative for HCV RNA for a long time and show positive antibody test results in the absence of circulating virus. Although antibody reactivity declines over time after spontaneous resolution of infection, T-cell responses might be maintained. A previous study (Semmo et al., 2005) has shown that in the absence of HCV RNA, 50% RIBA-indeterminate blood donors had HCV-specific T-cell responses, as determined by sensitive ex vivo analyses of T-cell response, and these responses were typically focused on core peptides. In another study, previous T-cell response to HCV recombinant proteins (core, NS3 and NS3 helicase) (Bes et al., 2009) was analysed using an interferon (IFN) enzyme-linked immunospot assay and the cytokines in culture supernatants
© 2014 British Blood Transfusion Society
Combination of two screening assays for HCV
301
Table 4. S/CO ratio distribution for eight screening EIAs for RIBA-positive samples EIAs S/CO
A
B
C
D
F
G
H
I
3·000
34 7 6
32 10 5
35 8 4
31 12 4
40 5 2
42 1 4
37 7 3
37 8 2
Table 5. Results of screening assay combinations Combination type Best
Worst
1 A,
1 2 3 4 5 6 7 8 9 10
Primary assay1 Second assay1 Number of positive samples detected Detection rate (%) P-value2 Ratio per 100 000 donations B A B B A F E C H F
D D C E C H F G G G
28 26 23 22 22 13 13 13 12 7
55·57 55·32 48·94 46·91 46·81 28·66 27·66 27·66 25·53 14·89
0·05
3·66 4·04 4·62 4·81 4·81 6·55 6·55 6·55 6·74 7·70
B, C, D, E, F, G, H, I and J represent the same assays as those described for Table 1. in the detection rates of two-assay combinations with that either of individual assays put together (𝜒 2 test).
2 Differences
from donors with different RIBA results were quantified. The results confirmed that in the absence of HCV RNA, approximately half of RIBA-indeterminate donors have a resolved, previous HCV infection. These findings suggest that indeterminate donors should be counselled appropriately with respect to potential previous HCV exposure. Therefore, donations of HCV RNA-negative with discrepant anti-HCV results might transmit infection and could be significant for transfusion safety. In the present study, none of the 47 donations HCV-RIBA positive (5·7%, 47 of 822) was NAT reactive; these samples might represent the type of case discussed above. Since the development of third-generation assays, several commercial EIA or CIA kits for the detection of HCV antibodies in plasma have become available. These kits use recombinant proteins containing linear epitopes from both structural and non-structural regions of HCV proteins (Lin et al., 2005). However, these kits do not always perform equally well. This may be because of a large degree of variation in the individual’s immune response to the different antigens used in the kits or the different vectors used to clone the various recombinant proteins (Zachary et al., 2005). Although most anti-HCV immunoassay kits contain NS3, NS4, NS5 and core recombinant antigens, some of the recombinant antigens could not react with antibodies’ immunologic response to real donor samples. A previous study (Couroucé et al., 1995) has shown that no current HCV screening test is able to detect all infectious and antibody-positive specimens in spite of technological improvements. The previous results obtained from seven anti-HCV screening tests, including Ortho, Sorin and
© 2014 British Blood Transfusion Society
Murex, show different numbers of false-negative results. In the present study, of the 47 RIBA-positive sera without detectable RNA, 10 screening assays for anti-HCV also yielded discordant results (detection rates ranging from 10·64 to 34·04%) that were unsatisfactory, and most reactive samples went undetected by any one kit. Thus, the issue regarding undetected samples could not be resolved by using just one screening assay in parallel with an HCV NAT. The missed detection of these 47 samples of donated blood could have an impact on the safety of blood transfusion. The second anti-HCV screening test could offer an alternative for supplementing or confirming results, and the UK Health Protection Agency has published an algorithm that advised use of a second EIA for confirmation (National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C: 2002 June 10-12, 2002). In 2013, US CDC issued an updated guidance (Centers for Disease Control and Prevention (CDC), 2013) for laboratory testing and reporting of HCV antibodies owing to new developments, including the discontinuation of RIBA, which has been used for supplemental testing of HCV antibodies. The guidance recommended using a second HCV antibody assay different from the first, as a supplement test for HCV antibody. The strategy of using a combination of two screening assays for anti-HCV detection has been validated in a routine clinical laboratory (Allain et al., 1996) and evaluated in healthy blood donors in early 1996 (Evaluations and Standards Laboratory 2005). These results mainly focused on the confirmation of true reactive sera and focused on avoiding incorrect exclusion of healthy blood donors; however, reactive
Transfusion Medicine, 2014, 24, 297–304
302 K. Zhang et al. 2 screening test for anti-HCV
All reactive
HCV Ab+
Discardb
All negative
One negative One reactive
HCV Ab-
HCV Ab IND
Perform HCV NAT
Nonreactive
Report
Discard1, 2
Reactive
Discard
Fig. 2. Proposed algorithm for HCV testing in blood centres based on anti-HCV antibodies1 These donated blood samples should be removed from the blood supply or future donations from these donors should not be accepted, should be further confirmed.2 For these donors, additional testing as appropriate should be performed according to updated guidance from the CDC [Centers for Disease Control and Prevention (CDC), 2013].
sera could not be avoided, especially seropositive samples, without detectable RNA. It has been demonstrated (Zachary et al., 2005; Vermeersch et al., 2008) that using a confirmation assay with a different kit, developed by a different manufacturer will reduce the risk of both tests detecting the same false-reactivity. The present study showed that the combined usage of any two anti-HCV screening assays could improve the detection rate of RIBA-positive sera without detectable RNA. Using the best combination of two screening assays for anti-HCV detection, the detection rates of 47 samples were increased from 46·81 to 55·57% compared with that of individual assays (𝜒 2 test, P < 0·05). The incidence rates for these true HCV-seropositive samples detected by 10 assays (5·97 to 8·09 of 100 000) were reduced by using the best combination of Sorin and Lizhu EIAs (3·66 per 100 000). This showed that two screening assays in parallel with an HCV NAT could improve the safety of blood transfusion. In the present study, eight RIBA patterns were obtained for the 47 positive samples. The sample numbers in four of them were fewer ( 0·05). S/CO ratio distribution for eight EIAs showed no significant characteristics (Table 4). The relatively small number of positive samples was a limitation of this study. In addition, our study found that even the best three rounds of screening assays (Ortho, Lizhu and Sorin EIAs) with NAT could
Transfusion Medicine, 2014, 24, 297–304
not increase the detection rate significantly. No difference in detection rate was found between the best combination of three (68·83%) or two (55·57%) tests (𝜒 2 test, P > 0·05). Thus, there is no evidence suggesting that the safety of blood transfusion was increased by the combination of three or more anti-HCV kits; such combinations tend to increase the cost of screening. Thus, according to the results of this study, even with two screening assays in addition to NAT, the detection rate for the 47 samples tested still did not reach 100% (as shown in Table 2, the highest detection rate of the 47 confirmed positive samples was 55·57% by Sorin and Lizhu tests). That would translate into 19 RIBA confirmed reactive samples of 519 229 (3·66 of 100 000) still missed by two rounds of screening assays with NAT. It was shown here that there is no suitable combination to achieve 100% sensitivity rate to avoid viral transmission. Thus, undetected samples could not be avoided, and the sensitivity of commercial anti-HCV kits should be improved. Finally, our proposed algorithm for HCV testing in blood centres based on anti-HCV results is summarised in Fig. 2. Compared with algorithms proposed by previous reports (Chapko et al., 2005; Vermeersch et al., 2008; Lai et al., 2011) using single anti-HCV screening tests, our proposed screening for anti-HCV combines two screening assays. A similar sequential immunoassay strategy in conjunction with HCV NAT screening was study in Australian donors, where donors whose samples were reactive in the primary screening immunoassay were tested with a secondary immunoassay, and if found to be reactive using both assays, the samples were further tested by immunoblot (Kiely et al., 2004). The Chiron RIBA HCV 3·0 Strip Immunoblot Assay (Chiron Corp, Emeryville, CA, USA) that was recommended in
© 2014 British Blood Transfusion Society
Combination of two screening assays for HCV this study for supplemental testing of blood samples after initial HCV antibody testing is no longer available and the only other FDA-approved supplemental tests for HCV infection are those that detect HCV viremia (CDC, 2013). Therefore, in our proposed algorithm, if the results of both immunoassays are reactive, then the sample is regarded as anti-HCV positive and no further confirmation is necessary; the sample should be discarded in this case. Even if the result of one of the two assays is positive, the sample should be discarded. At this stage, it should be further confirmed whether these samples of donated blood should be removed from the blood supply and whether future donations from these donors should not be accepted as well. For these donors, additional testing as appropriate should be performed according to updated guidance by the CDC (CDC, 2013). On the other hand, if the results of both assays are negative, then the sample should be tested with NAT to assess the presence of HCV RNA. This approach would improve the detection rate of reactive samples and increase the safety of donated blood. In conclusion, based on our study, the use of a single NAT for HCV RNA and a single anti-HCV immunoassay may be
REFERENCES Allain, J.P., Kitchen, A., Aloysius, S., Petrik, J., Barbara, J.A. & Williamson, L.M. (1996) Safety and efficacy of hepatitis C virus antibody screening of blood donors with two sequential screening assays. Transfusion, 36, 401–405. Bes, M., Esteban, J.I., Casamitjana, N. et al. (2009) Hepatitis C virus (HCV)-specific T-cell responses among recombinant immunoblot assay-3-indeterminate blood donors: a confirmatory evidence of HCV exposure. Transfusion, 49, 296–305. Busch, M.P., Glynn, S.A., Stramer, S.L. et al. (2005) A new strategy for estimating risks of transfusion-transmitted viral infections based on rates of detection of recently infected donors. Transfusion, 45, 254–264. Busch, M.P., Glynn, S.A., Stramer, S.L. et al. (2006) Correlates of hepatitis C virus (HCV) RNA negativity among HCV-seropositive blood donors. Transfusion, 46, 469–475. Centers for Disease Control and Prevention (1991) Public Health Service inter-agency guidelines for screening donors of blood, plasma, organs, tissues, and semen for evidence of hepatitis B and hepatitis C. MMWR - Recommendations and Reports, 40, 1–17. Centers for Disease Control and Prevention (1998) Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCV-related chronic disease. MMWR, 47, 1–33.
© 2014 British Blood Transfusion Society
303
insufficient to detect hepatitis C antibodies in blood samples that are non-reactive in the NAT. The combination of one NAT and two anti-HCV screening immunoassays is therefore of considerable importance for safety of blood donation.
ACKNOWLEDGMENTS This work was supported by the Chinese Ministry of Health Special Fund for Public Welfare (201002005) and the National High Technology Research and Development Program (863 Program) (2011AA02A116). We gratefully acknowledge all listed blood centres for the provision of samples. K. Z. and L. W. performed the research, J. L. designed the research study, Y. S. and R. Z. contributed essential reagents and equipments, J. X. and G. L. analysed the data, and K. Z. wrote the paper.
CONFLICT OF INTEREST The authors declare that they have no conflicts of interest relevant to the manuscript submitted to Transfusion Med.
Centers for Disease Control and Prevention (CDC) (2013) Testing for HCV infection: an update of guidance for clinicians and laboratorians, 62, 362-365. Chapko, M.K., Sloan, K.L., Davison, J.W., Dufour, D.R., Bankson, D.D., Rigsby, M., Dominitz, J.A. & Hepatitis C Resource Center Program, Department of Veterans Affairs. (2005) Cost effectiveness of testing strategies for chronic hepatitis C. American Journal of Gastroenterology, 100, 607–615. Contreras, A.M. (2006) Hepatitis C antibody: true or false? New diagnosis strategies. Revista de Investigación Clínica, 58, 153–160. Courouce, A.M., Le Marrec, N., Girault, A., Ducamp, S. & Simon, N. (1994) Anti-hepatitis C virus (anti-HCV) seroconversion in patients undergoing hemodialysis: comparison of secondand third-generation anti-HCV assays. Transfusion, 34, 790–795. Couroucé, A.M., Barin, F., Botté, C., Lunel, F., Maisonneuve, P., Maniez, M. & Trepo, C. (1995) A comparative evaluation of the sensitivity of seven anti-hepatitis C virus screening tests. Vox Sanguinis, 69, 213–216. Department of Health and Human Services Food and Drug Administration Center for Biologics Evaluation and Research (2004) Use of Nucleic Acid Tests on Pooled and Individual Samples from Donors of Whole Blood and Blood Components (including Source Plasma and Source Leukocytes) to Adequately and Appropriately Reduce the
Risk of Transmission of HIV-1 and HCV. U.S. Food and Drug Administration. Haydon, G.H., Jarvis, L.M., Blair, C.S., Simmonds, P., Harrison, D.J., Simpson, K.J. & Hayes, P.C. (1998) Clinical significance of intrahepatitis C virus levels in patients with chronic HCV infection. Gut, 42, 570–575. Houghton, M., Weiner, A., Han, J., Kuo, G. & Choo, Q.L. (1991) Molecular biology of the hepatitis C viruses: implications for diagnosis, development and control of viral disease. Hepatology, 14, 381–388. Kamili, S., Drobeniuc, J., Araujo, A.C. & Hayden, T.M. (2012) Laboratory diagnostics for hepatitis C virus infection. Clinical Infectious Diseases, 55 (Suppl. 1), S43–S548. Kiely, P., Kay, D., Parker, S. & Piscitelli, L. (2004) The significance of third-generation HCV RIBA-indeterminate, RNA-negative results in voluntary blood donorsscreened with sequential third-generation immunoassays. Transfusion, 44, 349–358. Kim, M.J., Park, Q., Min, H.K. & Kim, H.O. (2012) Residual risk of transfusiontransmitted infection with human immunodeficiency virus, hepatitis C virus, and hepatitis B virus in Korea from 2000 through 2010. BMC Infectious Diseases, 12, 160. Kleinman, S.H., Kuhns, M.C., Todd, D.S., Glynn, S.A., McNamara, A., DiMarco, A. & Busch, M.P., Retrovirus Epidemiology Donor Study. (2003) Retrovirus Epidemiology Donor Study. Frequency of HBV
Transfusion Medicine, 2014, 24, 297–304
304 K. Zhang et al. DNA detection in US blood donors testing positive for the presence of anti-HBc: implications for transfusion transmission and donor screening. Transfusion, 43, 696–704. Lai, K.K., Jin, M., Yuan, S., Larson, M.F., Dominitz, J.A., & Bankson, D.D. (2011) Improved reflexive testing algorithm for hepatitis C infection using signal-to-cutoff ratios of a hepatitis C virus antibody assay. Clinical Chemistry, 57, 1050–1056. Lanotte, P., Dubois, F., Le Pogam, S. et al. (1998) The kinetics of antibodies against hepatitis C virus may predict viral clearance in exposed hemophiliacs. Journal of Infectious Diseases, 178, 556–559. Lefrère, J.J., Girot, R., Lefrère, F., Guillaume, N., Lerable, J., Le Marrec, N., Bouchardeau, F. & Laperche, S. (2004) Complete or partial seroreversion in immunocompetent individuals after self-limited HCV infection: consequences for transfusion. Transfusion, 44, 343–348. Lin, S., Arcangel, P., Medina-Selby, A. et al. (2005) Design of novel conformational and genotype-specific antigens for improving sensitivity of immunoassays for hepatitis C virus-specific antibodies. Journal of Clinical Microbiology, 43, 3917–3924. Messick, K., Sanders, J.C., Goedert, J.J. & Eyster, M.E. (2001) Hepatitis C viral clearance and antibody reactivity patterns in
Transfusion Medicine, 2014, 24, 297–304
persons with haemophilia and other cogenital bleeding disorders. Haemophilia, 7, 568–574. National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C: 2002 June 10-12 (2002) Hepatology, 36, S3–S20. URL http://www3.interscience.wiley.com/cgi-bin /fulltext/106597945/PDFSTART (Accessed 2014/09/05). Nkrumah, B., Owusu, M. & Averu, P. (2011) Hepatitis B and C viral infections among blood donors. A retrospective study from a rural community of Ghana. BMC Research Notes, 12, 4529. Scheiblauer, H., El-Nageh, M., Nick, S. et al. (2006) Evaluation of the performance of 44 assays used in countries with limited resources for the detection of antibodies to hepatitis C virus. Transfusion, 46, 708–718. Seeff, L.B., Miller, R.N., Rabkin, C.S. et al. (2000) 45-year follow-up of hepatitis C virus infection in healthy young adults. Annals of Internal Medicine, 132, 105–111. Selvarajah, S. & Busch, M.P. (2012) Transfusion transmission of HCV, a long but successful road map to safety. Antiviral Therapy, 17 (7 Pt B),, 1423–1429. Semmo, N., Barnes, E., Taylor, C., Kurtz, J., Harcourt, G., Smith, N. & Klenerman, P.
(2005) T-cell responses and previous exposure to hepatitis C virus in indeterminate blood donors. Lancet, 365, 327–329. Stramer, S.L., Glynn, S.A., Kleinman, S.H. et al. (2004) Detection of HIV-1 and HCV infections among antibody-negative blood donors by nucleic acid amplification testing. New England Journal of Medicine, 351, 760–768. Strobl, F. (2011) NAT in blood screening around the world. MLO: Medical Laboratory Observer, 43, 12–14, 16, 18; quiz 20–21. Vermeersch, P., Van Ranst, M. & Lagrou, K. (2008) Validation of a strategy for HCV antibody testing with two enzyme immunoassays in a routine clinical laboratory. Journal of Clinical Virology, 42, 394–398. World Health Organization 2013. Hepatitis C Fact Sheet No. 164. (updated April 2014). URL http://www.who.int/mediacentre/ factsheets/fs164/en/ (Accessed 2014/09/05). Zachary, P., Ullmann, M., Djeddi, S., Meyer, N., Wendling, M.J., Schvoerer, E., Stoll-Keller, F. & Gut, J.P. (2005) Evaluation of three commercially available hepatitis C virus antibody detection assays under the conditions of a clinical virology laboratory. Journal of Clinical Virology, 34, 207–210; discussion 216-218.
© 2014 British Blood Transfusion Society
