- Email: [email protected]
An alternative strategy to western blot as a confirmatory diagnostic test for HIV infection
Accepted Manuscript Title: An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection Author: Xia Feng Jibao Wang Zhiyun Gao Yu Tian Ling Zhang Huichao Chen Tong Zhang Lin Xiao Jun Yao Wenge Xing Maofeng Qiu Yan Jiang PII: DOI: Reference:
S1386-6532(16)30638-2 http://dx.doi.org/doi:10.1016/j.jcv.2016.12.010 JCV 3750
To appear in:
Journal of Clinical Virology
Received date: Revised date: Accepted date:
25-8-2016 13-12-2016 20-12-2016
Please cite this article as: Feng Xia, Wang Jibao, Gao Zhiyun, Tian Yu, Zhang Ling, Chen Huichao, Zhang Tong, Xiao Lin, Yao Jun, Xing Wenge, Qiu Maofeng, Jiang Yan.An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection.Journal of Clinical Virology http://dx.doi.org/10.1016/j.jcv.2016.12.010 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection
Xia Fenga,b, Jibao Wangc, Zhiyun Gaod, Yu Tiana, Ling Zhanga, Huichao Chene, Tong Zhangb, Lin Xiaof, Jun Yaoa, Wenge Xinga, Maofeng Qiua, Yan Jianga*
a
National HIV/HCV Reference Laboratory, National Center for AIDS/STD Control
and Prevention, China CDC, Beijing, China b
Department of Clinical Laboratory, Beijing Youan Hospital, Capital Medical
University, Beijing, China c
Center for Disease Control and Prevention in Dehong, Dehong, China
d
Department of Pathogenic Biology, Hebei Medical University, Shijiazhuang, China
e
Center for Disease Control and Prevention in Yunnan Province, Kunming, China
f
Center for Disease Control and Prevention in Liangshan, Liangshan, China
*Corresponding author:
Yan Jiang, National HIV/HCV Reference Laboratory, National Center for AIDS/STD Control and Prevention, China CDC, Beijing, China Telephone number: +86-13911096250; Fax number: 010-58900994 Email: [email protected] 1
Highlights
The DBS–urine paired ELISA is equivalent to WB in HIV antibody detection.
The DBS–urine paired ELISA would be valuable in resource-limited regions.
The DBS–urine paired ELISA is affordable and easy to use.
Abstract Background In China, western blot (WB) is the recommended procedure for the diagnosis of HIV infection. However, this technique is time consuming and labor intensive, and its complexity restricts wide application in resource-limited regions. Objective The aim of this study was to evaluate the efficacy of a dry blood spots (DBS)–urine paired enzyme-linked immunosorbent assay (ELISA) test, instead of WB, for HIV antibody detection. Study design Plasma, DBS, and urine samples were collected from 1213 subjects from different populations. Two diagnostic testing strategies were conducted in parallel. The equivalence of the paired ELISA and WB strategies was assessed. Results A diagnosis of HIV was determined in 250 subjects according to the paired ELISA 2
test, and in 249 according to the WB strategy. The discordant case was judged HIV-positive during follow-up. In total, 18 subjects were diagnosed with possible HIV using the paired ELISA test, among whom, 11 subjects tested negative with WB, and one was confirmed to be HIV-positive during follow-up. For the remaining 945 subjects, both strategies indicated a negative result. The kappa test indicated good conformity (kappa = 0.954) between the two diagnostic strategies. Conclusion The DBS–urine paired ELISA could be applied as an alternative to WB in HIV diagnosis, which would be valuable in resource-limited regions owing to the associated affordability and ease of use.
Key words: ELISA; HIV; substitutive strategy; western blot 1. Background The Joint United Nations Programme on HIV/AIDS (UNAIDS) 90-90-90 program has become the common global target to help eradicate acquired immune deficiency syndrome (AIDS). The first 90 is the most important; by 2020, 90% of all people living with HIV will know their HIV status [1]. As far as HIV detection is concerned, it can meet the needs of detection when the window period is 14 days with the fourth-generation ELISA and 10 days with the nucleic acid test (NAT) [2]. However, in 2014, it was estimated that only 45% of all people with HIV were aware of their sero-status [3]. This low level may be related to the disparate capacities of laboratories, resulting from varying regional economies, and among key populations, 3
it may be associated with the fear of disclosing personal information and discrimination. Therefore, human immunodeficiency virus (HIV) diagnostic strategies have to be improved in both depth and breadth to satisfy the unmet needs. In China, the recommended test for confirmatory diagnosis of HIV infection is western blot (WB), which is widely used in the HIV testing network. For decades, this method has played an important role in the prevention and control of HIV. However, along with the development of new diagnostic methods, the popularity of WB faces several challenges such as a high rate of uncertainty, low sensitivity, and a long testing period, which could result in false-negatives for patients within the window period [4-8], and may also impede patients’ follow-up[8,9]. Moreover, in addition to regional differences in ethnicity, economy, and population, there is an uneven distribution of not only the prevalence of HIV infection but also the competency of HIV testing laboratories. Therefore, a single strategy is inadequate for HIV diagnosis. New diagnostic strategies with improved sensitivity, convenience, and regional applicability are needed for HIV/AIDS prevention and control. 2. Objectives In this study, the dry blood spot (DBS) and urine ELISA tests were combined as an alternative to WB. The practicability of the paired ELISA test was evaluated by direct comparison of these two strategies. 3. Study Design 3.1 Ethical considerations This study design was approved by the institution review board, and the 4
requirement for informed consent was waived. 3.2 Subjects From March 21, 2014 to January 31, 2015, a total of 1213 subjects were enrolled. This population included 261 subjects who were randomly selected from a population receiving a health examination by the disease prevention and control center (CDC), Liangshan Yi Autonomous Prefecture, Sichuan Province; 101 subjects with an HIV infection diagnosis confirmed by WB, with unknown transmission routes, sampled by Xichang sexually transmitted disease (STD) and skin disease prevention station; 251 drug users sampled by Dehong Prefecture CDC, Yunnan Province; and 600 men who have sex with men (MSM) sampled by Yunnan CDC and the outpatient service, Center for Infective Disease, Beijing Youan Hospital, Capital Medical University. 2.3 Sample processing Blood and urine samples were collected from each subject and numbered. Venous blood samples (3–5 mL) were centrifuged (3000 rpm/min), and serum was stored at 4°C. Urine samples (20–50 mL) were collected in special tubes that contain polymers to protect the antibody against degradation during transportation, and were stored at 4°C. 3.4 Reagents The urine HIV-1 antibody ELISA kits were provided by Beijing JunHe Pharmaceuticals Co., Ltd (Beijing, China). The blood HIV antibody ELISA kits were provided by Beijing Wantai Biological Pharmacy (Beijing, China). The WB kits for HIV antibody were provided by MP Biomedicals Asia Pacific Pte., Ltd (Singapore). 5
The NATs were performed for nucleic acid detection using the COBAS Ampli Prep/COBAS Taq Man HIV-1 Test (version 2.0) (Swiss). Internal quality control for the urine ELISA was formulated by mixing HIV-antibody positive plasma and HIV-antibody negative urine with signal to cutoff (S/CO) ratios of 3–5. Internal quality control for the blood ELISA was provided by Beijing Kinghawk Pharmaceutical Co., Ltd. (Beijing, China). 3.5 Testing procedure After sampling, all tests were performed within seven days by the same technician from the National Reference Laboratory, National Center for AIDS/STD Control and Prevention, China CDC. Samples from Dehong Prefecture were tested on site, but all of the other samples were transported to the National Reference Laboratory for testing. Blood samples were successively screened, retested, and confirmed in accordance with the National Guidelines for HIV Diagnosis (revised in 2009). Sero-positive and WB-negative samples were tested for HIV nucleic acid. For sero-negative samples, 50-sample pooled NATs were performed[10]. 3.6 DBS–urine paired ELISA test Urine and DBS samples were experiment as an alternative strategy，also screened by ELISA. An HIV positive diagnosis was indicated by positive results from both the DBS and urine samples in the parallel ELISA tests An HIV negative diagnosis was indicated by double negative results. In cases where either the DBS or urine ELISA indicated a positive result, “possible” HIV was diagnosed and further follow-up was required(Figure 1).. 6
3.7 Statistical analysis Concordance between blood ELISA, urine ELISA, WB, and NAT results was calculated,and the conformity between the paired ELISA and WB tests was evaluated using Kappa value. Data analyses were performed by using SPSS 19.0 (SPSS Inc., Chicago, IL) and p<0.05 was considered statistically significant. 4. Results For all 1213 subjects, the paired ELISA results were consistent with those of the WB blood analysis.Figure 2 presents a comparison of the WB and paired ELISA tests. In total, 250 subjects were confirmed to be HIV positive by the DBS–urine paired ELISA. Of these, 249 subjects were also positive on the WB. One subject had possible HIV with a band present at gp160 and a plasma viral load of 3.5 x 10 6 copies/mL. After one-month follow-up, the result for this case was positive with bands observed at p24, gp120, and gp160. These test results combined with clinical manifestations and epidemiological history indicated a diagnosis of early HIV infection. Overall, a positive result from the paired ELISA test was a good indicator of HIV infection. For the remaining 945 subjects both strategies indicated a negative result, confirmed by negative pooled NATs. There were 18 (1.48%) cases of “possible” HIV identified by the DBS–urine paired ELISA. Of the nine cases where DBS was positive and urine was negative, six screened positive with WB. The other three cases were negative by WB. Corresponding viral load tests indicated that, of the three cases, two were negative 7
and one had a viral load of 25 x 106 copies/mL. After 49 days follow-up, a repeated WB test showed a positive result with bands present at p24, gp41, gp120, and gp160 for this subject. Indicating that, for a sample taken within the window period, a false-negative is more likely with WB than the alternative testing strategy. In contrast, for the nine cases where urine was positive and DBS was negative, the WB results were all negative. The S/CO ratios for the urine ELISA were all low (1.052–2.032). These results suggest a potential for false-positive results with the urine antibody detection test. However, the World Health Organization (WHO) report that a positive urine HIV antibody test may indicate an HIV infection before blood HIV antibodies can be detected [11]. The cross-sectional design and lack of follow-up in our study mean that we cannot assess whether the positive urine tests are an early indication of infection in these cases. Therefore, further research is needed to investigate the value of urine antibody tests as early indicators of HIV infection. The sensitivity and specificity of the paired ELISA strategy were 97.6% (249/255) and 98.7%(945/957) when taking WB as the gold standard, respectively. The total concordance between the two diagnostic testing strategies was 98.43% (1194/1213), with good consistency (Kappa = 0.954 > 0.85, P = 0.000) (Figure 2). Direct comparison of the paired ELISA with WB shows that there is very good consistency between the two strategies.
5. Discussion The UNAIDS guidelines for using testing technologies in HIV surveillance state 8
that ELISA and rapid diagnostic tests (RDTs) are the most precise and cost-effective methods [15],and the use of ELISA and RDTs for HIV detection has been widely adopted internationally, including in the United States and in most European countries [16-20]. At present, only a few countries, such as China and France, use WB for the detection of HIV. Furthermore, WB faces several challenges such as a high rate of uncertainty, low sensitivity, and a long testing period which leads to the loss of patients. On the other hand, China’s use of WB for HIV diagnosis confirmation is contrary to the recommendations in the WHO’s 2015 consolidated guidelines on HIV testing services [15]. Therefore, China needs to reevaluate their testing strategies. A parallel screening technique using two blood ELISA tests from different manufacturers has been used in the blood donor screening program in China for more than 30 years, which has proven that practical application of a parallel ELISA method is feasible [21]. Since it was first approved by the US Food and Drug Administration in 1996, DBS has proved to be an efficient way of collecting samples and has equivalent accuracy to blood. DBS is now used in many countries in HIV surveillance [22-25]. In China, XiHui Zang [26] demonstrated equivalence between DBS and blood sampling methods. The same conclusion can be drawn from our study. Our research group has also demonstrated that excellent results can be gained through using urine samples for HIV antibody detection [27]. The sensitivity and specificity of urine HIV antibody testing are reported to be 69.7%~97% and 99%~100%, respectively [28,29], which are in line with our results. 9
When the paired ELISA with WB testing strategies wre compsred (table 2), for WB, the experimental process is time consuming, technical requirements are high, and interpretation of the results is difficult. Therefore, for remote and underdeveloped areas, samples need to be collected and submitted to a specialist HIV laboratory, which not only increases the burden on the HIV laboratory but also increases the turnaround time for the test results potentially leading to losses. This is the first study to evaluate an alternative method for HIV antibody detection using two biological samples and a parallel ELISA. This technique can not only identify cases of HIV infection, but also can get confirmatory results in a shorter time using ELISA. Furthermore, it is suitable for mass sampling. Moreover, this parallel strategy flags up possible infections that require follow-up and/or further diagnostic tests, which may reduce missed diagnosis of cases tested within the window period.
6. Conclusions In conclusion, the DBS–urine paired ELISA test can be applied as an alternative strategy for HIV diagnosis in remote and resource-limited areas. It may also be valuable for home-based self-testing in high-risk populations who do not want to be tested because of fears of discrimination. All above, this alternative strategy can increase the accessibility of HIV testing. The sample size in this study was limited, which may bias the results. Therefore, future research needs to be designed to evaluate the feasibility of the paired ELISA 10
strategy in different regions and in populations with different AIDS incidence rates.
Funding sources The work was supported by the National Science and Technology Major Project of the 12th Five-Year Plan (2013ZX10001001-001-007) and the Beijing Municipal Science and technology project (D141100000314003). The authors have no conflicts of interest to declare. Acknowledgements The authors would like to acknowledge all of the staff from the participating laboratories. We would also like to thank Mengjie Bai for providing helpful feedback on an earlier version of the paper.
11
References [1] UNAIDS, 90-90-90. An ambitious treatment target to help end the AIDS epidemic, Joint United Nations Programme on HIV/AIDS (UNAIDS), 2014. [2] N.E. Rosenberg, C.D. Pilcher, M.P. Busch, M.S. Cohen, How can we better identify early HIV infections? Curr. Opin. HIV AIDS 10 (2015) 61–68. [3] AIDS JUNPoH, The gap report. UNAIDS, Geneva, 2014. [4] X.J. Zhang, L. Zhang, X. Li, Analysis of the tracking and monitoring results in patients with uncertain HIV-1 antibody, Chin. J. Health Lab. Tec. 14 (2015) 2396–2398. [5] J. Peng, J. He, Y. Jiang, The analysis on indeterminate results of HIV antibodies from 2008 to 2012 in Hunan Province, Chin. J. AIDS STD 4 (2014) 240–243. [6] C. Liuqing, H. Xuena, Z. Danyan, The follow-up of HIV antibody undetermined, Zhejiang Prev. Med. 1 (2015) 68–70. [7] I. Yongyan, W. Fangfang, F. Changhua, N. Xiansheng, H. Fenglan H, Analysis of 645 confirmatory test cases of HIV antibody screening positive samples, Pract. Prev. Med. 3 (2014) 368–369. [8] Z. Yalan, Z. Haichao, Y. Yang, Y. Tongtong, W. Xiaoli, Analysis of misdetection and its causes in WB assay of HIV confirmatory test, Chin. J. AIDS STD 5 (2014) 309–311. [9] C.Y. Liang, Z.G. Han, Q.S. Chen, Assessment of 103 cases which were positive by fourth-generation combined HIV antigen/antibody ELISA screening test and negative by WB confirmatory test, Chin. J. AIDS STD 17 (2011) 19–21. 12
[10] J. Huazhou, S. Sheng, P. Lijian, H. Xiaojie, W. Hao, et al, Evaluation of application of pooling nucleic acid amplification testing in men who have sex with men population in China, Chin. J. Lab. Med. 34 (2011) 529–533. [11] W. Oelemann, D. Verissimo, G. Costa, et al. Diagnostic detection of human immunodeficiency virus type 1 antibodies in urine: a Brazilian study. J. Clin. Microbiol. 40 (2002) 881–885. [12] F. Xia, Y. Jing, D. Yong,
Epidemiological analysis of HIV/AIDS patients in
Yunnan provincial general hospital. China Modern Doctor 2 (2013) 17–19. [13] UNAIDS/WHO, Working group on global HIV/AIDS/ATT surveillance. Guidelines for using HIV testing technologies in surveillance, 2009. [14] Y.H. Ran, Study on the time measurement and unit cost analysis of HIV counseling and testing, Chinese Center for Disease Control and Prevention, 2008. [15] WHO, Consolidated guidelines on HIV testing services, 2015. [16] Haute Autorité de Santé, HIV infection screening in France laboratory tests and algorithms, 2008. [17] US Centers for Disease Control and Prevention, Laboratory testing for the diagnosis of HIV infection, 2014. [18] Wisconsin State Laboratory of Hygiene. New HIV-1/2 antibody testing algorithm implemented HIV-1 Western Blot test discontinued, 2013. [19] D. Gokengin, A.M. Geretti, J. Begovac, A. Palfreeman, M. Stevanovic, O. Tarasenko, K. Radcliffe, 2014 European guideline on HIV testing, Int. J. STD AIDS 25 (2014) 695–704. 13
[20] Y.P. Yin, E. Ngige, C. Anyaike, G. Ijaodola, T.A. Oyelade, R.G. Vaz, L.M. Newman, X.S. Chen, Laboratory evaluation of three dual rapid diagnostic tests for HIV and syphilis in China and Nigeria, Int. J. Gynaecol. Obstet. 130 (2015) S22–S26. [21] P. Zeng, J. Liu, J. Wang. Parallel enzyme-linked immunosorbent assay screening for human immunodeficiency virus among blood donors in five Chinese blood centres: a retrospective analysis. Transfus Med. 25 (2015) 259–264. [22] WHO, Report on the first international symposium on self-testing for HIV: the legal, ethical, gender, human rights and public health implication of HIV self-testing scale-up, Geneva, April 8–9, 2013. [23] S.S. Solomon, S. Solomon, I.I. Rodriguez, S.T. McGarvey, A.K. Ganesh, S.P. Thyagarajan, A.P. Mahajan, K.H. Mayer., Dried blood spots (DBS): a valuable tool for HIV surveillance in developing/tropical countries, Int. J. STD AIDS 13 (2002) 25–28. [24] E.T. Montgomery, B. van der Pol, A. van der Straten, G. Ramjee, G. de Bruyn, T. Chipato, K. Blanchard, N.S. Padian; MIRA Team, Discrepancies in diagnosis of incident HIV infection between antibody- and DNA-based tests in a phase III prevention trial in southern Africa, Int. J. STD AIDS 23 (2012) 649–652. [25] D. Kania, A.M. Bekalé, N. Nagot, A.M. Mondain, L. Ottomani, N. Meda, M. Traoré, J.B. Ouédraogo, J. Ducos, P. van de Perre, E. Tuaillon, Combining rapid diagnostic tests and dried blood spot assays for point-of-care testing of human immunodeficiency virus, hepatitis B and hepatitis C infections in Burkina Faso, 14
West Africa, Clin. Microbiol. Infect. 19 (2013) E533–541. [26] Z. Yanxi, Z. Hela, Z. Qi, L. Qiurong, Q. Laiying, Z. Guiyun, et al., Study on HIV Antibody Detection by Paper with Dried Blood Spot(DBS) Specimens. Chin J AIDS STD. 11 (2005) 323–325. [27] F. Xia,W. Jibao, T. Yu, Clinical evaluation of urine HIV-1 antibody tests. Chin J AIDS STD, 2016 (4)241–243. [28] C. Holm-Hansen, N.T. Constantine, N. Haukenes, Detection of antibodies to HIV in homologous sets of plasma, urine and oral mucosal transudate samples using rapid assays in Tanzania, Clin. Diagn. Virol. 1 (1993) 207–214. [29] P. xianwu, S. zhiyong, W. lu. Comparative study on the results of HIV-1 antibody detection in urine and blood samples. Chin J AIDS STD[J], 2016 (2) 126–128.
15
Figure 1. Parallel DBS and urine samples in the paired ELISA
Abbreviations: D= dry blood spots; U=urine
Figure 2. Sample processing and qualitative results
Abbreviations: DBS = dry blood spots; WB = western blot *One subject had possible HIV with a band present at gp160 and a plasma viral load of 3.5 x 106 copies/mL. After one-month follow-up, the result for this case was positive with bands observed at p24, gp120, and gp160.
16
Table 1. Comparison of the results of the WB and paired ELISA strategies WB Total
Paired ELISA
+
Possible
-
+
249
1
0
250
Possible
6
0
12
18
-
0
0
945
945
255
1
957
1213
Total
Abbreviations: WB = western blot; ELISA = enzyme-linked immunosorbent assay. The total concordance between the two diagnostic testing strategies was 98.43% (1194/1213), Kappa = 0.954 , P = 0.000
Table 2. Comparison of the paired ELISA and WB testing strategies WB
Paired ELISA
Laboratory requirement
Confirmatory lab
Screening and confirmatory lab
Strategy design
Serial
Parallel
Operational convenience
Complicated
Easy
Testing period (days)
3~7[12,13]
1~2
Expense (RMB)
207.8~408.7 [14]
50~100
Ratio of uncertainty (%)
3.03~14.5 [5-8]
1.48
Possibility of false-negatives
High
Low
Abbreviations: WB = western blot; ELISA = enzyme-linked immunosorbent assay.
17
S1386-6532(16)30638-2 http://dx.doi.org/doi:10.1016/j.jcv.2016.12.010 JCV 3750
To appear in:
Journal of Clinical Virology
Received date: Revised date: Accepted date:
25-8-2016 13-12-2016 20-12-2016
Please cite this article as: Feng Xia, Wang Jibao, Gao Zhiyun, Tian Yu, Zhang Ling, Chen Huichao, Zhang Tong, Xiao Lin, Yao Jun, Xing Wenge, Qiu Maofeng, Jiang Yan.An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection.Journal of Clinical Virology http://dx.doi.org/10.1016/j.jcv.2016.12.010 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection An Alternative Strategy to Western Blot as a Confirmatory Diagnostic Test for HIV Infection
Xia Fenga,b, Jibao Wangc, Zhiyun Gaod, Yu Tiana, Ling Zhanga, Huichao Chene, Tong Zhangb, Lin Xiaof, Jun Yaoa, Wenge Xinga, Maofeng Qiua, Yan Jianga*
a
National HIV/HCV Reference Laboratory, National Center for AIDS/STD Control
and Prevention, China CDC, Beijing, China b
Department of Clinical Laboratory, Beijing Youan Hospital, Capital Medical
University, Beijing, China c
Center for Disease Control and Prevention in Dehong, Dehong, China
d
Department of Pathogenic Biology, Hebei Medical University, Shijiazhuang, China
e
Center for Disease Control and Prevention in Yunnan Province, Kunming, China
f
Center for Disease Control and Prevention in Liangshan, Liangshan, China
*Corresponding author:
Yan Jiang, National HIV/HCV Reference Laboratory, National Center for AIDS/STD Control and Prevention, China CDC, Beijing, China Telephone number: +86-13911096250; Fax number: 010-58900994 Email: [email protected] 1
Highlights
The DBS–urine paired ELISA is equivalent to WB in HIV antibody detection.
The DBS–urine paired ELISA would be valuable in resource-limited regions.
The DBS–urine paired ELISA is affordable and easy to use.
Abstract Background In China, western blot (WB) is the recommended procedure for the diagnosis of HIV infection. However, this technique is time consuming and labor intensive, and its complexity restricts wide application in resource-limited regions. Objective The aim of this study was to evaluate the efficacy of a dry blood spots (DBS)–urine paired enzyme-linked immunosorbent assay (ELISA) test, instead of WB, for HIV antibody detection. Study design Plasma, DBS, and urine samples were collected from 1213 subjects from different populations. Two diagnostic testing strategies were conducted in parallel. The equivalence of the paired ELISA and WB strategies was assessed. Results A diagnosis of HIV was determined in 250 subjects according to the paired ELISA 2
test, and in 249 according to the WB strategy. The discordant case was judged HIV-positive during follow-up. In total, 18 subjects were diagnosed with possible HIV using the paired ELISA test, among whom, 11 subjects tested negative with WB, and one was confirmed to be HIV-positive during follow-up. For the remaining 945 subjects, both strategies indicated a negative result. The kappa test indicated good conformity (kappa = 0.954) between the two diagnostic strategies. Conclusion The DBS–urine paired ELISA could be applied as an alternative to WB in HIV diagnosis, which would be valuable in resource-limited regions owing to the associated affordability and ease of use.
Key words: ELISA; HIV; substitutive strategy; western blot 1. Background The Joint United Nations Programme on HIV/AIDS (UNAIDS) 90-90-90 program has become the common global target to help eradicate acquired immune deficiency syndrome (AIDS). The first 90 is the most important; by 2020, 90% of all people living with HIV will know their HIV status [1]. As far as HIV detection is concerned, it can meet the needs of detection when the window period is 14 days with the fourth-generation ELISA and 10 days with the nucleic acid test (NAT) [2]. However, in 2014, it was estimated that only 45% of all people with HIV were aware of their sero-status [3]. This low level may be related to the disparate capacities of laboratories, resulting from varying regional economies, and among key populations, 3
it may be associated with the fear of disclosing personal information and discrimination. Therefore, human immunodeficiency virus (HIV) diagnostic strategies have to be improved in both depth and breadth to satisfy the unmet needs. In China, the recommended test for confirmatory diagnosis of HIV infection is western blot (WB), which is widely used in the HIV testing network. For decades, this method has played an important role in the prevention and control of HIV. However, along with the development of new diagnostic methods, the popularity of WB faces several challenges such as a high rate of uncertainty, low sensitivity, and a long testing period, which could result in false-negatives for patients within the window period [4-8], and may also impede patients’ follow-up[8,9]. Moreover, in addition to regional differences in ethnicity, economy, and population, there is an uneven distribution of not only the prevalence of HIV infection but also the competency of HIV testing laboratories. Therefore, a single strategy is inadequate for HIV diagnosis. New diagnostic strategies with improved sensitivity, convenience, and regional applicability are needed for HIV/AIDS prevention and control. 2. Objectives In this study, the dry blood spot (DBS) and urine ELISA tests were combined as an alternative to WB. The practicability of the paired ELISA test was evaluated by direct comparison of these two strategies. 3. Study Design 3.1 Ethical considerations This study design was approved by the institution review board, and the 4
requirement for informed consent was waived. 3.2 Subjects From March 21, 2014 to January 31, 2015, a total of 1213 subjects were enrolled. This population included 261 subjects who were randomly selected from a population receiving a health examination by the disease prevention and control center (CDC), Liangshan Yi Autonomous Prefecture, Sichuan Province; 101 subjects with an HIV infection diagnosis confirmed by WB, with unknown transmission routes, sampled by Xichang sexually transmitted disease (STD) and skin disease prevention station; 251 drug users sampled by Dehong Prefecture CDC, Yunnan Province; and 600 men who have sex with men (MSM) sampled by Yunnan CDC and the outpatient service, Center for Infective Disease, Beijing Youan Hospital, Capital Medical University. 2.3 Sample processing Blood and urine samples were collected from each subject and numbered. Venous blood samples (3–5 mL) were centrifuged (3000 rpm/min), and serum was stored at 4°C. Urine samples (20–50 mL) were collected in special tubes that contain polymers to protect the antibody against degradation during transportation, and were stored at 4°C. 3.4 Reagents The urine HIV-1 antibody ELISA kits were provided by Beijing JunHe Pharmaceuticals Co., Ltd (Beijing, China). The blood HIV antibody ELISA kits were provided by Beijing Wantai Biological Pharmacy (Beijing, China). The WB kits for HIV antibody were provided by MP Biomedicals Asia Pacific Pte., Ltd (Singapore). 5
The NATs were performed for nucleic acid detection using the COBAS Ampli Prep/COBAS Taq Man HIV-1 Test (version 2.0) (Swiss). Internal quality control for the urine ELISA was formulated by mixing HIV-antibody positive plasma and HIV-antibody negative urine with signal to cutoff (S/CO) ratios of 3–5. Internal quality control for the blood ELISA was provided by Beijing Kinghawk Pharmaceutical Co., Ltd. (Beijing, China). 3.5 Testing procedure After sampling, all tests were performed within seven days by the same technician from the National Reference Laboratory, National Center for AIDS/STD Control and Prevention, China CDC. Samples from Dehong Prefecture were tested on site, but all of the other samples were transported to the National Reference Laboratory for testing. Blood samples were successively screened, retested, and confirmed in accordance with the National Guidelines for HIV Diagnosis (revised in 2009). Sero-positive and WB-negative samples were tested for HIV nucleic acid. For sero-negative samples, 50-sample pooled NATs were performed[10]. 3.6 DBS–urine paired ELISA test Urine and DBS samples were experiment as an alternative strategy，also screened by ELISA. An HIV positive diagnosis was indicated by positive results from both the DBS and urine samples in the parallel ELISA tests An HIV negative diagnosis was indicated by double negative results. In cases where either the DBS or urine ELISA indicated a positive result, “possible” HIV was diagnosed and further follow-up was required(Figure 1).. 6
3.7 Statistical analysis Concordance between blood ELISA, urine ELISA, WB, and NAT results was calculated,and the conformity between the paired ELISA and WB tests was evaluated using Kappa value. Data analyses were performed by using SPSS 19.0 (SPSS Inc., Chicago, IL) and p<0.05 was considered statistically significant. 4. Results For all 1213 subjects, the paired ELISA results were consistent with those of the WB blood analysis.Figure 2 presents a comparison of the WB and paired ELISA tests. In total, 250 subjects were confirmed to be HIV positive by the DBS–urine paired ELISA. Of these, 249 subjects were also positive on the WB. One subject had possible HIV with a band present at gp160 and a plasma viral load of 3.5 x 10 6 copies/mL. After one-month follow-up, the result for this case was positive with bands observed at p24, gp120, and gp160. These test results combined with clinical manifestations and epidemiological history indicated a diagnosis of early HIV infection. Overall, a positive result from the paired ELISA test was a good indicator of HIV infection. For the remaining 945 subjects both strategies indicated a negative result, confirmed by negative pooled NATs. There were 18 (1.48%) cases of “possible” HIV identified by the DBS–urine paired ELISA. Of the nine cases where DBS was positive and urine was negative, six screened positive with WB. The other three cases were negative by WB. Corresponding viral load tests indicated that, of the three cases, two were negative 7
and one had a viral load of 25 x 106 copies/mL. After 49 days follow-up, a repeated WB test showed a positive result with bands present at p24, gp41, gp120, and gp160 for this subject. Indicating that, for a sample taken within the window period, a false-negative is more likely with WB than the alternative testing strategy. In contrast, for the nine cases where urine was positive and DBS was negative, the WB results were all negative. The S/CO ratios for the urine ELISA were all low (1.052–2.032). These results suggest a potential for false-positive results with the urine antibody detection test. However, the World Health Organization (WHO) report that a positive urine HIV antibody test may indicate an HIV infection before blood HIV antibodies can be detected [11]. The cross-sectional design and lack of follow-up in our study mean that we cannot assess whether the positive urine tests are an early indication of infection in these cases. Therefore, further research is needed to investigate the value of urine antibody tests as early indicators of HIV infection. The sensitivity and specificity of the paired ELISA strategy were 97.6% (249/255) and 98.7%(945/957) when taking WB as the gold standard, respectively. The total concordance between the two diagnostic testing strategies was 98.43% (1194/1213), with good consistency (Kappa = 0.954 > 0.85, P = 0.000) (Figure 2). Direct comparison of the paired ELISA with WB shows that there is very good consistency between the two strategies.
5. Discussion The UNAIDS guidelines for using testing technologies in HIV surveillance state 8
that ELISA and rapid diagnostic tests (RDTs) are the most precise and cost-effective methods [15],and the use of ELISA and RDTs for HIV detection has been widely adopted internationally, including in the United States and in most European countries [16-20]. At present, only a few countries, such as China and France, use WB for the detection of HIV. Furthermore, WB faces several challenges such as a high rate of uncertainty, low sensitivity, and a long testing period which leads to the loss of patients. On the other hand, China’s use of WB for HIV diagnosis confirmation is contrary to the recommendations in the WHO’s 2015 consolidated guidelines on HIV testing services [15]. Therefore, China needs to reevaluate their testing strategies. A parallel screening technique using two blood ELISA tests from different manufacturers has been used in the blood donor screening program in China for more than 30 years, which has proven that practical application of a parallel ELISA method is feasible [21]. Since it was first approved by the US Food and Drug Administration in 1996, DBS has proved to be an efficient way of collecting samples and has equivalent accuracy to blood. DBS is now used in many countries in HIV surveillance [22-25]. In China, XiHui Zang [26] demonstrated equivalence between DBS and blood sampling methods. The same conclusion can be drawn from our study. Our research group has also demonstrated that excellent results can be gained through using urine samples for HIV antibody detection [27]. The sensitivity and specificity of urine HIV antibody testing are reported to be 69.7%~97% and 99%~100%, respectively [28,29], which are in line with our results. 9
When the paired ELISA with WB testing strategies wre compsred (table 2), for WB, the experimental process is time consuming, technical requirements are high, and interpretation of the results is difficult. Therefore, for remote and underdeveloped areas, samples need to be collected and submitted to a specialist HIV laboratory, which not only increases the burden on the HIV laboratory but also increases the turnaround time for the test results potentially leading to losses. This is the first study to evaluate an alternative method for HIV antibody detection using two biological samples and a parallel ELISA. This technique can not only identify cases of HIV infection, but also can get confirmatory results in a shorter time using ELISA. Furthermore, it is suitable for mass sampling. Moreover, this parallel strategy flags up possible infections that require follow-up and/or further diagnostic tests, which may reduce missed diagnosis of cases tested within the window period.
6. Conclusions In conclusion, the DBS–urine paired ELISA test can be applied as an alternative strategy for HIV diagnosis in remote and resource-limited areas. It may also be valuable for home-based self-testing in high-risk populations who do not want to be tested because of fears of discrimination. All above, this alternative strategy can increase the accessibility of HIV testing. The sample size in this study was limited, which may bias the results. Therefore, future research needs to be designed to evaluate the feasibility of the paired ELISA 10
strategy in different regions and in populations with different AIDS incidence rates.
Funding sources The work was supported by the National Science and Technology Major Project of the 12th Five-Year Plan (2013ZX10001001-001-007) and the Beijing Municipal Science and technology project (D141100000314003). The authors have no conflicts of interest to declare. Acknowledgements The authors would like to acknowledge all of the staff from the participating laboratories. We would also like to thank Mengjie Bai for providing helpful feedback on an earlier version of the paper.
11
References [1] UNAIDS, 90-90-90. An ambitious treatment target to help end the AIDS epidemic, Joint United Nations Programme on HIV/AIDS (UNAIDS), 2014. [2] N.E. Rosenberg, C.D. Pilcher, M.P. Busch, M.S. Cohen, How can we better identify early HIV infections? Curr. Opin. HIV AIDS 10 (2015) 61–68. [3] AIDS JUNPoH, The gap report. UNAIDS, Geneva, 2014. [4] X.J. Zhang, L. Zhang, X. Li, Analysis of the tracking and monitoring results in patients with uncertain HIV-1 antibody, Chin. J. Health Lab. Tec. 14 (2015) 2396–2398. [5] J. Peng, J. He, Y. Jiang, The analysis on indeterminate results of HIV antibodies from 2008 to 2012 in Hunan Province, Chin. J. AIDS STD 4 (2014) 240–243. [6] C. Liuqing, H. Xuena, Z. Danyan, The follow-up of HIV antibody undetermined, Zhejiang Prev. Med. 1 (2015) 68–70. [7] I. Yongyan, W. Fangfang, F. Changhua, N. Xiansheng, H. Fenglan H, Analysis of 645 confirmatory test cases of HIV antibody screening positive samples, Pract. Prev. Med. 3 (2014) 368–369. [8] Z. Yalan, Z. Haichao, Y. Yang, Y. Tongtong, W. Xiaoli, Analysis of misdetection and its causes in WB assay of HIV confirmatory test, Chin. J. AIDS STD 5 (2014) 309–311. [9] C.Y. Liang, Z.G. Han, Q.S. Chen, Assessment of 103 cases which were positive by fourth-generation combined HIV antigen/antibody ELISA screening test and negative by WB confirmatory test, Chin. J. AIDS STD 17 (2011) 19–21. 12
[10] J. Huazhou, S. Sheng, P. Lijian, H. Xiaojie, W. Hao, et al, Evaluation of application of pooling nucleic acid amplification testing in men who have sex with men population in China, Chin. J. Lab. Med. 34 (2011) 529–533. [11] W. Oelemann, D. Verissimo, G. Costa, et al. Diagnostic detection of human immunodeficiency virus type 1 antibodies in urine: a Brazilian study. J. Clin. Microbiol. 40 (2002) 881–885. [12] F. Xia, Y. Jing, D. Yong,
Epidemiological analysis of HIV/AIDS patients in
Yunnan provincial general hospital. China Modern Doctor 2 (2013) 17–19. [13] UNAIDS/WHO, Working group on global HIV/AIDS/ATT surveillance. Guidelines for using HIV testing technologies in surveillance, 2009. [14] Y.H. Ran, Study on the time measurement and unit cost analysis of HIV counseling and testing, Chinese Center for Disease Control and Prevention, 2008. [15] WHO, Consolidated guidelines on HIV testing services, 2015. [16] Haute Autorité de Santé, HIV infection screening in France laboratory tests and algorithms, 2008. [17] US Centers for Disease Control and Prevention, Laboratory testing for the diagnosis of HIV infection, 2014. [18] Wisconsin State Laboratory of Hygiene. New HIV-1/2 antibody testing algorithm implemented HIV-1 Western Blot test discontinued, 2013. [19] D. Gokengin, A.M. Geretti, J. Begovac, A. Palfreeman, M. Stevanovic, O. Tarasenko, K. Radcliffe, 2014 European guideline on HIV testing, Int. J. STD AIDS 25 (2014) 695–704. 13
[20] Y.P. Yin, E. Ngige, C. Anyaike, G. Ijaodola, T.A. Oyelade, R.G. Vaz, L.M. Newman, X.S. Chen, Laboratory evaluation of three dual rapid diagnostic tests for HIV and syphilis in China and Nigeria, Int. J. Gynaecol. Obstet. 130 (2015) S22–S26. [21] P. Zeng, J. Liu, J. Wang. Parallel enzyme-linked immunosorbent assay screening for human immunodeficiency virus among blood donors in five Chinese blood centres: a retrospective analysis. Transfus Med. 25 (2015) 259–264. [22] WHO, Report on the first international symposium on self-testing for HIV: the legal, ethical, gender, human rights and public health implication of HIV self-testing scale-up, Geneva, April 8–9, 2013. [23] S.S. Solomon, S. Solomon, I.I. Rodriguez, S.T. McGarvey, A.K. Ganesh, S.P. Thyagarajan, A.P. Mahajan, K.H. Mayer., Dried blood spots (DBS): a valuable tool for HIV surveillance in developing/tropical countries, Int. J. STD AIDS 13 (2002) 25–28. [24] E.T. Montgomery, B. van der Pol, A. van der Straten, G. Ramjee, G. de Bruyn, T. Chipato, K. Blanchard, N.S. Padian; MIRA Team, Discrepancies in diagnosis of incident HIV infection between antibody- and DNA-based tests in a phase III prevention trial in southern Africa, Int. J. STD AIDS 23 (2012) 649–652. [25] D. Kania, A.M. Bekalé, N. Nagot, A.M. Mondain, L. Ottomani, N. Meda, M. Traoré, J.B. Ouédraogo, J. Ducos, P. van de Perre, E. Tuaillon, Combining rapid diagnostic tests and dried blood spot assays for point-of-care testing of human immunodeficiency virus, hepatitis B and hepatitis C infections in Burkina Faso, 14
West Africa, Clin. Microbiol. Infect. 19 (2013) E533–541. [26] Z. Yanxi, Z. Hela, Z. Qi, L. Qiurong, Q. Laiying, Z. Guiyun, et al., Study on HIV Antibody Detection by Paper with Dried Blood Spot(DBS) Specimens. Chin J AIDS STD. 11 (2005) 323–325. [27] F. Xia,W. Jibao, T. Yu, Clinical evaluation of urine HIV-1 antibody tests. Chin J AIDS STD, 2016 (4)241–243. [28] C. Holm-Hansen, N.T. Constantine, N. Haukenes, Detection of antibodies to HIV in homologous sets of plasma, urine and oral mucosal transudate samples using rapid assays in Tanzania, Clin. Diagn. Virol. 1 (1993) 207–214. [29] P. xianwu, S. zhiyong, W. lu. Comparative study on the results of HIV-1 antibody detection in urine and blood samples. Chin J AIDS STD[J], 2016 (2) 126–128.
15
Figure 1. Parallel DBS and urine samples in the paired ELISA
Abbreviations: D= dry blood spots; U=urine
Figure 2. Sample processing and qualitative results
Abbreviations: DBS = dry blood spots; WB = western blot *One subject had possible HIV with a band present at gp160 and a plasma viral load of 3.5 x 106 copies/mL. After one-month follow-up, the result for this case was positive with bands observed at p24, gp120, and gp160.
16
Table 1. Comparison of the results of the WB and paired ELISA strategies WB Total
Paired ELISA
+
Possible
-
+
249
1
0
250
Possible
6
0
12
18
-
0
0
945
945
255
1
957
1213
Total
Abbreviations: WB = western blot; ELISA = enzyme-linked immunosorbent assay. The total concordance between the two diagnostic testing strategies was 98.43% (1194/1213), Kappa = 0.954 , P = 0.000
Table 2. Comparison of the paired ELISA and WB testing strategies WB
Paired ELISA
Laboratory requirement
Confirmatory lab
Screening and confirmatory lab
Strategy design
Serial
Parallel
Operational convenience
Complicated
Easy
Testing period (days)
3~7[12,13]
1~2
Expense (RMB)
207.8~408.7 [14]
50~100
Ratio of uncertainty (%)
3.03~14.5 [5-8]
1.48
Possibility of false-negatives
High
Low
Abbreviations: WB = western blot; ELISA = enzyme-linked immunosorbent assay.
17