- Email: [email protected]
Evaluation of a domestic interferon-gamma release assay for detecting Mycobacterium tuberculosis infection in China
Tuberculosis 95 (2015) 523e526
Contents lists available at ScienceDirect
Tuberculosis journal homepage: http://intl.elsevierhealth.com/journals/tube
Evaluation of a domestic interferon-gamma release assay for detecting Mycobacterium tuberculosis infection in China Yongliang Liu a, b, 1, Mingzhan Ou a, b, 1, Shuizhen He b, c, Xiaofei Li d, Yanyan Lin a, b, Junhui Xiong b, Jun Zhang a, b, Shengxiang Ge a, b, * a
State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, School of Public Health, Xiamen University, Xiamen, Fujian, 361102, PR China National Institute of Diagnostics and Vaccine Development in Infectious Diseases, School of Life Science, Xiamen University, Xiamen, Fujian, 361102, PR China c Xiamen Center for Disease Control and Prevention, Xiamen, Fujian, 361021, PR China d Third People's Hospital of Kunming City, Kunming, Yunnan, 650041, PR China b
a r t i c l e i n f o
s u m m a r y
Article history: Received 4 November 2014 Accepted 15 May 2015
Interferon-gamma release assays (IGRAs) have been demonstrated to be useful in the diagnosis of Mycobacterium tuberculosis (MTB) infection. However, IGRAs have not been recommended for clinical usage in most low-income countries due to the shortage of clinical data available resulting from their high test cost. Recently, a cheaper domestic TB-IGRA was approved in China. In this study, we compared TB-IGRA with QuantiFERON-TB Gold In-Tube (QFT-GIT) for MTB infection diagnosis in 253 active TB patients, 48 non-TB lung disease patients, 115 healthcare workers and 216 healthy individuals. The proportion of positive TB-IGRA results in active TB patients, patients with non-TB lung disease, healthcare workers and healthy individuals was 88.3%, 27.1%, 40.9% and 17.6%, respectively, which was similar to the results of QFT-GIT, with an overall agreement of 95% (k ¼ 0.89) and a high correlation between their responses (r ¼ 0.85, p < 0.001) being observed. In conclusion, the TB-IGRA has comparable clinical performance with QFT-GIT. © 2015 Elsevier Ltd. All rights reserved.
Keywords: IGRAs TB-IGRA Performance MTB infection
1. Introduction Tuberculosis (TB), caused by Mycobacterium tuberculosis (MTB), remains a global threat to public health with approximately 8.6 million new cases and 1.3 million deaths in 2012 [1]. It is estimated that one-third of the world's population is infected with MTB, and the majority have a latent infection. Persons with latent TB infection (LTBI) are clinically asymptomatic but have a 5%e10% lifetime risk that latent mycobacteria will become active and cause TB [2]. LTBI screening and prophylactic treatment can substantially reduce the risk of the development of disease and are important TB control approaches [3]. MTB infection can elicit a robust adaptive cell-mediated immune response, which has been employed to identify MTB infection, especially LTBI. There are two approaches currently used to
* Corresponding author. Xiamen University, South Xiang'an Road, Xiamen, 361102, PR China. Tel./fax: þ86 592 2181258. E-mail address: [email protected] (S. Ge). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.tube.2015.05.007 1472-9792/© 2015 Elsevier Ltd. All rights reserved.
determine the adaptive immunity, the tuberculin skin test (TST) and the gamma interferon (IFN-g) release assay (IGRA). The TST measures delayed-type hypersensitivity reactions to a crude mixture of MTB antigens which are also present in bacillus Calmrin (BCG) and non-tuberculous mycobacteria (NTM). IGRA ette-Gue is based on measurement of IFN-g secreted from T cells previously exposed to MTB when stimulated in vitro with the MTB-specific antigens, such as ESAT-6 and CFP-10. Both antigens are encoded by RD1, a genomic region present in M. tuberculosis but lacking in all Mycobacterium bovis BCG vaccine strains and most of the NTM [4]. There are two commercial IGRAs currently available, the QuantiFERON-TB Gold In-Tube test (QFT-GIT) (Cellestis, Carnegie, Australia) and the T-SPOT.TB assay (Oxford Immunotec, Oxford, UK). With higher sensitivity and specificity for detecting MTB infection compared with TST, IGRAs have been widely used to diagnose MTB infection under national guidelines in many developed countries, such as the USA, UK and Japan [5]. However, in most developing countries, including China, the clinical utilization of IGRAs is not recommended due to insufficient evidence of their
524
Y. Liu et al. / Tuberculosis 95 (2015) 523e526
performance in high TB burden settings [6,7]. Hitherto, the available clinical data are too limited to formulate guidelines for how IGRAs are used in the clinical situation in China. The high test cost of QFT-GIT and T-SPOT.TB is one of restrictive factors of their clinical usage in developing countries [8]. Recently, in China, a domestic TB IFN-g release assay (TB-IGRA) (Beijing Wantai, Beijing, China), which is cheaper than the two imported IGRAs, QFT-GIT and T-SPOT.TB, was licensed by the China Food and Drug Administration (CFDA). The objective of this study is to evaluate the performance of TB-IGRA in populations at various levels of risk for MTB infection, compared with the QFT-GIT in China. A high quality and cheap assay will be beneficial to the clinical utilization of IGRA and accumulating more clinical data for demonstrating its clinical value in developing countries with high TB burden. 2. Results and discussion The diagnosis of MTB infection is very important to control TB epidemics, such as the investigation and control of TB outbreaks [9], surveillance of high risk populations [10,11], and epidemiological study of LTBI [12,13]. Because there is no diagnostic gold standard for MTB infection, it is difficult to accurately evaluate the performance of an assay to diagnose MTB infection, especially LTBI. In low-incidence settings, active TB patients and low-risk individuals with no known TB exposure are often used as surrogates of confirmed positive and negative MTB infected populations to estimate the assay's sensitivity and specificity, respectively [14]. However, in high-incidence settings, such as China, there is no suitable population that can be considered as the surrogate of subjects without MTB infection, due to the relatively high frequency of TB exposure in the population. One alternate method is to investigate whether the assay's results in different populations reflected their risks of MTB infection. In this study, three populations with different MTB infection risks were recruited to evaluate the performance of TB-IGRA for MTB infection diagnosis, active TB patients with confirmed MTB infection, healthcare workers (HCWs) with increased MTB infection risk due to frequent TB exposure, and healthy individuals with normal risk. From September 2010 to July 2011, a total of 331 suspected active TB patients, 115 HCWs working in TB wards and 216 healthy individuals visiting the Center of Physical Examination were recruited in the Third People's Hospital of Kunming City, Kunming, Yunnan, China. Diagnosis of active TB among the recruited patients was made on the basis of all clinical, radiological, microbiological and histopathological information collected after recruitment and response to anti-TB therapy for at least 3 months. Among the 331 suspected TB patients, 255 active pulmonary TB patients (PTBs) and 28 extra-pulmonary TB patients (EPTBs) were eventually confirmed, and 48 were diagnosed with non-tuberculosis lung disease. The 28 EPTBs consisted of 14 cases of pleural tuberculosis, 7 cases of tuberculous peritonitis, 4 cases of nephrotuberculosis, 2
cases of spinal tuberculosis and 1 case of tubercular lymphadenitis. Of the 48 patients with non-tuberculosis lung disease, 35 were pneumonia, and the others included lung cancer (n ¼ 9), pneumoconiosis (n ¼ 2), chronic bronchitis (n ¼ 1) and hydrothorax (n ¼ 1). This study received ethical approval from the Ethics Committee of the School of Public Health, Xiamen University. Written Informed consent was obtained from each participant. The demographic characteristics of the subjects in this study are shown in Table 1. The number of male subjects was greater than the number of females in the populations of PTBs, EPTBs, nontuberculosis lung disease and healthy individuals, but more females were found among HCWs. More than half of HCWs were nurses (67 out of 115), all of whom were female. Of the HCWs, the working years in TB high risk situation were from 1 to 40, with the mean of 14.27 ± 9.40. At enrollment, all subjects were tested with TB-IGRA and QFT-GIT according to the manufacturer's instructions. Briefly, for TB-IGRA, one milliliter of fresh heparinized venous whole blood was added separately into each two-milliliter Eppendorf centrifuge tube containing nil for negative control (N), mitogen for positive control (P) and TB antigen (a recombinant fusion protein of CFP-10 and ESAT-6) (T). After hour incubation at 37 C, each tube was centrifuged, and the concentration of IFN-g in the plasma was measured using the ELISA method. The IFN-g value (pg/ml) for the TB antigen and mitogen were corrected for background by subtracting the value of N, namely T-N and P-N. As recommended by the manufacturer, the result of the test was interpreted as positive (N 400 pg/ml, T-N 14 pg/ml and 25% of N), negative (N 400 pg/ml and P-N 14 pg/ml, T-N < 20 pg/ ml or T-N 4 pg/ml but <25% of N) or indeterminate (N > 400 pg/ml, or N 400 pg/ml and P-N < 14 pg/ml). For QFTGIT, one millimeter of whole blood was collected separately in each heparin-containing tube pre-coated with nil for negative control, mitogen for positive control and TB antigen (peptides from ESAT-6, CFP-10 and TB7.7 [Rv2654c]). After 24 h incubation at 37 C, the sample tubes were centrifuged, and the plasma was collected for measuring the IFN-g concentration by ELISA. The result were interpreted as positive, negative or indeterminate on the basis of the manufacturer's recommended cutoff value (IFN-g 0.35 IU/ml) [15]. Valid results of TB-IGRA and QFT-GIT were available from all 662 subjects and are shown in Table 1. There were 6 indeterminate results, 1 for TB-IGRA and 5 for QFT-GIT. The positive rate of TBIGRA was 88.6%, 85.7%, 27.1%, 40.9% and 17.6% in PTBs, EPTBs, non-tuberculosis lung disease, HCWs and healthy individuals, respectively. Those were similar to the results of QFT-GIT with positive rates of 86.7%, 85.7%, 29.1%, 41.7% and 15.7% in the corresponding populations. The positive rates of the two IGRAs in different populations manifested a positive correlation with their risk of MTB infection. The positive rates were highest in the population diagnosed with active TB (88.3% of TB-IGRA and 86.6% of QFT-GIT), followed by the populations with increased MTB
Table 1 Characteristics of study population and the results of TB-IGRA and QFT-GIT. Population
No.
M/F*
Mean age ± SD (years)
TB-IGRA Positive (%)
Active TB patients PTBs EPTBs Non-TB lung disease HCWs Healthy individuals * y
283 255 28 48 115 216
182/101 164/91 18/10 28/20 40/75 122/94
M/F indicates the ratio of male/female. IND indicates the indeterminate results of IGRAs.
40.8 41.0 36.8 48.9 36.2 30.9
± ± ± ± ± ±
17.2 17.2 15.7 16.1 9.8 8.7
250 226 24 13 47 38
(88.3%) (88.6%) (85.7%) (27.1%) (40.9%) (17.6%)
QFT-GIT IND
1
y
Positive (%) 245 221 24 14 48 34
(86.6%) (86.7%) (85.7%) (29.2%) (41.7%) (15.7%)
IND 3 2
Y. Liu et al. / Tuberculosis 95 (2015) 523e526 Table 2 Agreement between the TB-IGRA and QFT-GIT results (n ¼ 656). TB-IGRA results þ QFT-GIT results þ 325 20 Total 345
Total
16 295 311
341 315 656
Agreement (95% CI)
Kappa value (95% CI)
95% (93%e96%)
0.89 (0.86e0.92)
infection risk (HCWs, 40.9% of TB-IGRA and 41.7% of QFT-GIT), and lowest in the populations with normal MTB infection risk (non-TB lung disease and healthy individuals) (27.1% and 17.6% of TB-IGRA, 29.2% and 15.7% of QFT-GIT). The results indicates that TB-IGRA has the comparable sensitivity and ability to reflect the risk levels of MTB infection in different populations with QFT-GIT, in another word, the two assays, TB-IGRA and QFT-GIT, have the same sensitivity and specificity. For further comparison of TB-IGRA and QFT-GIT, the result agreement and response correlation between the two tests were analyzed by SPSS 11.0 (SPSS Inc., Chicago, IL, USA) in 656 subjects without indeterminate results on any the two assays. Concordance between the tests was measured using the Kappa index (k value > 0.75, excellent agreement; 0.75 k 0.4, fair to good agreement; k < 0.4, poor agreement), and the correlation between the TB-IGRA response and the QFT-GIT response was analyzed nonparametrically by Spearman's correlation. All of the significance tests were two sided, and p < 0.05 was considered to be statistically significant. In the 656 subjects, both TB-IGRA and QFT-GIT were positive in 325 and negative in 295 subjects, and the observed agreement between the tests was 95%, with excellent concordance (k ¼ 0.89, 95% CI: 0.86e0.92) (Table 2). When investigating different populations, the agreement was 94%, 100%, 84%, 92% and 97% in PTBs, EPTBs, non-TB lung diseases, HCWs and healthy individuals, respectively (data not shown). The response correlation between the two assays was analyzed and described in Table 3. The IFN-g concentrations (T-N) of TB-IGRA were highly correlated to those of QFT-GIT in all subjects with r ¼ 0.85 (95% CI: 0.82e0.87; p < 0.001). The correlation coefficient in both positive subjects was 0.71 (95% CI: 0.65e0.76; p < 0.001), which was higher than that in both negative subjects (r ¼ 0.22, 95% CI: 0.82e0.87; p < 0.001). Although the correlation coefficient decreased from 0.81 in active TB patients to 0.47 in healthy individuals with the decrease of their positive rates, the responses of TB-IGRA were correlated significantly with those of QFT-GIT in all study populations with different risks of MTB infection (all p < 0.0001) (Table 3). These results indicate that there are excellent concordance between the results of TB-IGRA and QFT-GIT and good correlation between the responses of the two tests. In principle, there is no
Table 3 Correlations between the TB-IGRA responses and the QFT-GIT responses. Population (n)
Spearman r (95% CI)
p value
Overall subjects (656) Both positive (325) Both negative (295) Active TB patients (280) PTBs (252) EPTBs (28) Non-TB lung disease (45) HCWs (115) Healthy individuals (216)
0.85 0.71 0.22 0.81 0.81 0.81 0.67 0.72 0.47
<0.0001 <0.0001 0.0002 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
(0.82e0.87) (0.65e0.76) (0.10e0.33) (0.76e0.85) (0.76e0.85) (0.62e0.91) (0.46e0.81) (0.62e0.80) (0.35e0.57)
525
significant difference between the two tests. Both assays employ ELISA to measure the concentration of IFN-g released from lymphocytes in heparinized whole blood stimulated with MTB antigens compared with controls. The major difference between the two assays is the antigens used to stimulate the lymphocytes. TBIGRA uses a recombinant fusion protein of the region of difference-1 (RD1) antigens CFP-10 and ESTA-6 as the stimulus, whereas QFT-GIT uses pooled peptides from CFP-10 and ESAT-6, as well as an additional antigen TB7.7 (Rv2654c, which is not an RD1 antigen). But, previous studies have demonstrated that T-cell responses to recombinant antigens CFT-10 and ESTA-6 are similar to those to corresponding mixtures of synthetic overlapping peptides in active TB patients [16] and healthy subjects with latent TB infection [17]. In this study, most of discrepant results occurred in the individuals with low IFN-g response (see Supplementary Figure 1), which may be due to the decreased quantitative accuracy of IFN-g ELISA assays around their lower limit of detection rather than the difference of antigens used by TB-IGRA and QFT-GIT. 3. Conclusion In summary, TB-IGRA, a recently licensed Chinese domestic TB IFN-g release assay, has similar clinical performance to QFT-GIT in China. Due to this assay's low cost and high quality, it may be a useful tool in active TB diagnosis and MTB infection detection in low-income countries, where TB is endemic. Financial support: This work was supported by a grant from the Institute Reconstruction Fund (2011FU125Z04), Xiamen City Municipal platform fund on viral biotherapeutics (3502Z20131001) and National Science and Technology Major Project (2008ZX10003). Conflict of interest: We declare that we have no financial or personal relationships with other people or organizations that can inappropriately influence our work. There is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript. Ethical approval: This study received ethical approval from the Ethics Committee of the School of Public Health, Xiamen University. Written Informed consent was obtained from each participant. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.tube.2015.05.007. References [1] WHO. Global tuberculosis report 2013. 2013. [2] O'Garra A, Redford PS, McNab FW, Bloom CI, Wilkinson RJ, Berry MP. The immune response in tuberculosis. Annu Rev Immunol 2013;31:475e527. PubMed PMID: 23516984. [3] Dye C, Williams BG. Eliminating human tuberculosis in the twenty-first century. Journal of the Royal Society. Interface/R Soc 2008 Jun 6;5(23):653e62. PubMed PMID: 17690054. Pubmed Central PMCID: 3226985. [4] Gordon SV, Brosch R, Billault A, Garnier T, Eiglmeier K, Cole ST. Identification of variable regions in the genomes of tubercle bacilli using bacterial artificial chromosome arrays. Mol Microbiol 1999;32(3):643e55. [5] Denkinger C, Dheda K, Pai M. Guidelines on interferon-g release assays for tuberculosis infection: concordance, discordance or confusion? Clin Microbiol Infect 2011;17(6):806e14. [6] Metcalfe JZ, Everett CK, Steingart KR, Cattamanchi A, Huang L, Hopewell PC, et al. Interferon-gamma release assays for active pulmonary tuberculosis diagnosis in adults in low- and middle-income countries: systematic review
526
[7] [8]
[9]
[10]
[11]
[12]
Y. Liu et al. / Tuberculosis 95 (2015) 523e526 and meta-analysis. J Infect Dis 2011 Nov 15;204(Suppl. 4):S1120e9. PubMed PMID: 21996694. Pubmed Central PMCID: 3192542. WHO. Use of tuberculosis interferon-gamma release assays (IGRAs) in lowand middle income countries. 2011. ISBN 978 92 4 150267 2. Steffen RE, Caetano R, Pinto M, Chaves D, Ferrari R, Bastos M, et al. Costeffectiveness of Quantiferon(R)-TB Gold-in-Tube versus tuberculin skin testing for contact screening and treatment of latent tuberculosis infection in Brazil. PLoS One 2013;8(4):e59546. PubMed PMID: 23593145. Pubmed Central PMCID: 3617186. Song Q, Guo H, Zhong H, Liu Z, Chen X, Wang C, et al. Evaluation of a new interferon-gamma release assay and comparison to tuberculin skin test during a tuberculosis outbreak. Int J Infect Dis IJID Off Publ Int Soc Infect Dis 2012 Jul;16(7):e522e6. PubMed PMID: 22542006. Wolf T, Goetsch U, Oremek G, Bickel M, Khaykin P, Haberl A, et al. Tuberculosis skin test, but not interferon-gamma-releasing assays is affected by BCG vaccination in HIV patients. J Infect 2013 Apr;66(4):376e80. PubMed PMID: 23159357. Munoz L, Santin M. Interferon-gamma release assays versus tuberculin skin test for targeting people for tuberculosis preventive treatment: an evidencebased review. J Infect 2013 Apr;66(4):381e7. PubMed PMID: 23298892. Legesse M, Ameni G, Mamo G, Medhin G, Bjune G, Abebe F. Community-based cross-sectional survey of latent tuberculosis infection in Afar pastoralists, Ethiopia, using QuantiFERON-TB Gold In-Tube and tuberculin skin test. BMC
[13]
[14] [15] [16]
[17]
Infect Dis 2011;11:89. PubMed PMID: 21477326. Pubmed Central PMCID: 3080306. Mekaini LA, Jabri ON, Narchi H, Kamal SM, Mabrook A, Kuwaiti MM, et al. The use of an interferon-gamma release assay to screen for pediatric latent tuberculosis infection in the eastern region of the Emirate of Abu Dhabi. Int J Infect Dis IJID Off Publ Int Soc Infect Dis 2014 Jun;23:4e7. PubMed PMID: 24657274. Huebner RE, Schein MF, Bass Jr JB. The tuberculin skin test. Clin Infect Dis Off Publ Infect Dis Soc Am 1993 Dec;17(6):968e75. PubMed PMID: 8110954. Cellestis. QuantiFERON®-TB gold (In-Tube method) [package insert]. Carnegie, VIC, Australia: Cellestis; 2006. Arend SM, Geluk A, van Meijgaarden KE, van Dissel JT, Theisen M, Andersen P, et al. Antigenic equivalence of human T-cell responses to Mycobacterium tuberculosis-specific RD1-encoded protein antigens ESAT-6 and culture filtrate protein 10 and to mixtures of synthetic peptides. Infect Immun 2000 Jun;68(6):3314e21. PubMed PMID: 10816479. Pubmed Central PMCID: 97589. Arend SM, Engelhard AC, Groot G, de Boer K, Andersen P, Ottenhoff TH, et al. Tuberculin skin testing compared with T-cell responses to Mycobacterium tuberculosis-specific and nonspecific antigens for detection of latent infection in persons with recent tuberculosis contact. Clin Diagn Lab Immunol 2001 Nov;8(6):1089e96. PubMed PMID: 11687445. Pubmed Central PMCID: 96231.
Contents lists available at ScienceDirect
Tuberculosis journal homepage: http://intl.elsevierhealth.com/journals/tube
Evaluation of a domestic interferon-gamma release assay for detecting Mycobacterium tuberculosis infection in China Yongliang Liu a, b, 1, Mingzhan Ou a, b, 1, Shuizhen He b, c, Xiaofei Li d, Yanyan Lin a, b, Junhui Xiong b, Jun Zhang a, b, Shengxiang Ge a, b, * a
State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, School of Public Health, Xiamen University, Xiamen, Fujian, 361102, PR China National Institute of Diagnostics and Vaccine Development in Infectious Diseases, School of Life Science, Xiamen University, Xiamen, Fujian, 361102, PR China c Xiamen Center for Disease Control and Prevention, Xiamen, Fujian, 361021, PR China d Third People's Hospital of Kunming City, Kunming, Yunnan, 650041, PR China b
a r t i c l e i n f o
s u m m a r y
Article history: Received 4 November 2014 Accepted 15 May 2015
Interferon-gamma release assays (IGRAs) have been demonstrated to be useful in the diagnosis of Mycobacterium tuberculosis (MTB) infection. However, IGRAs have not been recommended for clinical usage in most low-income countries due to the shortage of clinical data available resulting from their high test cost. Recently, a cheaper domestic TB-IGRA was approved in China. In this study, we compared TB-IGRA with QuantiFERON-TB Gold In-Tube (QFT-GIT) for MTB infection diagnosis in 253 active TB patients, 48 non-TB lung disease patients, 115 healthcare workers and 216 healthy individuals. The proportion of positive TB-IGRA results in active TB patients, patients with non-TB lung disease, healthcare workers and healthy individuals was 88.3%, 27.1%, 40.9% and 17.6%, respectively, which was similar to the results of QFT-GIT, with an overall agreement of 95% (k ¼ 0.89) and a high correlation between their responses (r ¼ 0.85, p < 0.001) being observed. In conclusion, the TB-IGRA has comparable clinical performance with QFT-GIT. © 2015 Elsevier Ltd. All rights reserved.
Keywords: IGRAs TB-IGRA Performance MTB infection
1. Introduction Tuberculosis (TB), caused by Mycobacterium tuberculosis (MTB), remains a global threat to public health with approximately 8.6 million new cases and 1.3 million deaths in 2012 [1]. It is estimated that one-third of the world's population is infected with MTB, and the majority have a latent infection. Persons with latent TB infection (LTBI) are clinically asymptomatic but have a 5%e10% lifetime risk that latent mycobacteria will become active and cause TB [2]. LTBI screening and prophylactic treatment can substantially reduce the risk of the development of disease and are important TB control approaches [3]. MTB infection can elicit a robust adaptive cell-mediated immune response, which has been employed to identify MTB infection, especially LTBI. There are two approaches currently used to
* Corresponding author. Xiamen University, South Xiang'an Road, Xiamen, 361102, PR China. Tel./fax: þ86 592 2181258. E-mail address: [email protected] (S. Ge). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.tube.2015.05.007 1472-9792/© 2015 Elsevier Ltd. All rights reserved.
determine the adaptive immunity, the tuberculin skin test (TST) and the gamma interferon (IFN-g) release assay (IGRA). The TST measures delayed-type hypersensitivity reactions to a crude mixture of MTB antigens which are also present in bacillus Calmrin (BCG) and non-tuberculous mycobacteria (NTM). IGRA ette-Gue is based on measurement of IFN-g secreted from T cells previously exposed to MTB when stimulated in vitro with the MTB-specific antigens, such as ESAT-6 and CFP-10. Both antigens are encoded by RD1, a genomic region present in M. tuberculosis but lacking in all Mycobacterium bovis BCG vaccine strains and most of the NTM [4]. There are two commercial IGRAs currently available, the QuantiFERON-TB Gold In-Tube test (QFT-GIT) (Cellestis, Carnegie, Australia) and the T-SPOT.TB assay (Oxford Immunotec, Oxford, UK). With higher sensitivity and specificity for detecting MTB infection compared with TST, IGRAs have been widely used to diagnose MTB infection under national guidelines in many developed countries, such as the USA, UK and Japan [5]. However, in most developing countries, including China, the clinical utilization of IGRAs is not recommended due to insufficient evidence of their
524
Y. Liu et al. / Tuberculosis 95 (2015) 523e526
performance in high TB burden settings [6,7]. Hitherto, the available clinical data are too limited to formulate guidelines for how IGRAs are used in the clinical situation in China. The high test cost of QFT-GIT and T-SPOT.TB is one of restrictive factors of their clinical usage in developing countries [8]. Recently, in China, a domestic TB IFN-g release assay (TB-IGRA) (Beijing Wantai, Beijing, China), which is cheaper than the two imported IGRAs, QFT-GIT and T-SPOT.TB, was licensed by the China Food and Drug Administration (CFDA). The objective of this study is to evaluate the performance of TB-IGRA in populations at various levels of risk for MTB infection, compared with the QFT-GIT in China. A high quality and cheap assay will be beneficial to the clinical utilization of IGRA and accumulating more clinical data for demonstrating its clinical value in developing countries with high TB burden. 2. Results and discussion The diagnosis of MTB infection is very important to control TB epidemics, such as the investigation and control of TB outbreaks [9], surveillance of high risk populations [10,11], and epidemiological study of LTBI [12,13]. Because there is no diagnostic gold standard for MTB infection, it is difficult to accurately evaluate the performance of an assay to diagnose MTB infection, especially LTBI. In low-incidence settings, active TB patients and low-risk individuals with no known TB exposure are often used as surrogates of confirmed positive and negative MTB infected populations to estimate the assay's sensitivity and specificity, respectively [14]. However, in high-incidence settings, such as China, there is no suitable population that can be considered as the surrogate of subjects without MTB infection, due to the relatively high frequency of TB exposure in the population. One alternate method is to investigate whether the assay's results in different populations reflected their risks of MTB infection. In this study, three populations with different MTB infection risks were recruited to evaluate the performance of TB-IGRA for MTB infection diagnosis, active TB patients with confirmed MTB infection, healthcare workers (HCWs) with increased MTB infection risk due to frequent TB exposure, and healthy individuals with normal risk. From September 2010 to July 2011, a total of 331 suspected active TB patients, 115 HCWs working in TB wards and 216 healthy individuals visiting the Center of Physical Examination were recruited in the Third People's Hospital of Kunming City, Kunming, Yunnan, China. Diagnosis of active TB among the recruited patients was made on the basis of all clinical, radiological, microbiological and histopathological information collected after recruitment and response to anti-TB therapy for at least 3 months. Among the 331 suspected TB patients, 255 active pulmonary TB patients (PTBs) and 28 extra-pulmonary TB patients (EPTBs) were eventually confirmed, and 48 were diagnosed with non-tuberculosis lung disease. The 28 EPTBs consisted of 14 cases of pleural tuberculosis, 7 cases of tuberculous peritonitis, 4 cases of nephrotuberculosis, 2
cases of spinal tuberculosis and 1 case of tubercular lymphadenitis. Of the 48 patients with non-tuberculosis lung disease, 35 were pneumonia, and the others included lung cancer (n ¼ 9), pneumoconiosis (n ¼ 2), chronic bronchitis (n ¼ 1) and hydrothorax (n ¼ 1). This study received ethical approval from the Ethics Committee of the School of Public Health, Xiamen University. Written Informed consent was obtained from each participant. The demographic characteristics of the subjects in this study are shown in Table 1. The number of male subjects was greater than the number of females in the populations of PTBs, EPTBs, nontuberculosis lung disease and healthy individuals, but more females were found among HCWs. More than half of HCWs were nurses (67 out of 115), all of whom were female. Of the HCWs, the working years in TB high risk situation were from 1 to 40, with the mean of 14.27 ± 9.40. At enrollment, all subjects were tested with TB-IGRA and QFT-GIT according to the manufacturer's instructions. Briefly, for TB-IGRA, one milliliter of fresh heparinized venous whole blood was added separately into each two-milliliter Eppendorf centrifuge tube containing nil for negative control (N), mitogen for positive control (P) and TB antigen (a recombinant fusion protein of CFP-10 and ESAT-6) (T). After hour incubation at 37 C, each tube was centrifuged, and the concentration of IFN-g in the plasma was measured using the ELISA method. The IFN-g value (pg/ml) for the TB antigen and mitogen were corrected for background by subtracting the value of N, namely T-N and P-N. As recommended by the manufacturer, the result of the test was interpreted as positive (N 400 pg/ml, T-N 14 pg/ml and 25% of N), negative (N 400 pg/ml and P-N 14 pg/ml, T-N < 20 pg/ ml or T-N 4 pg/ml but <25% of N) or indeterminate (N > 400 pg/ml, or N 400 pg/ml and P-N < 14 pg/ml). For QFTGIT, one millimeter of whole blood was collected separately in each heparin-containing tube pre-coated with nil for negative control, mitogen for positive control and TB antigen (peptides from ESAT-6, CFP-10 and TB7.7 [Rv2654c]). After 24 h incubation at 37 C, the sample tubes were centrifuged, and the plasma was collected for measuring the IFN-g concentration by ELISA. The result were interpreted as positive, negative or indeterminate on the basis of the manufacturer's recommended cutoff value (IFN-g 0.35 IU/ml) [15]. Valid results of TB-IGRA and QFT-GIT were available from all 662 subjects and are shown in Table 1. There were 6 indeterminate results, 1 for TB-IGRA and 5 for QFT-GIT. The positive rate of TBIGRA was 88.6%, 85.7%, 27.1%, 40.9% and 17.6% in PTBs, EPTBs, non-tuberculosis lung disease, HCWs and healthy individuals, respectively. Those were similar to the results of QFT-GIT with positive rates of 86.7%, 85.7%, 29.1%, 41.7% and 15.7% in the corresponding populations. The positive rates of the two IGRAs in different populations manifested a positive correlation with their risk of MTB infection. The positive rates were highest in the population diagnosed with active TB (88.3% of TB-IGRA and 86.6% of QFT-GIT), followed by the populations with increased MTB
Table 1 Characteristics of study population and the results of TB-IGRA and QFT-GIT. Population
No.
M/F*
Mean age ± SD (years)
TB-IGRA Positive (%)
Active TB patients PTBs EPTBs Non-TB lung disease HCWs Healthy individuals * y
283 255 28 48 115 216
182/101 164/91 18/10 28/20 40/75 122/94
M/F indicates the ratio of male/female. IND indicates the indeterminate results of IGRAs.
40.8 41.0 36.8 48.9 36.2 30.9
± ± ± ± ± ±
17.2 17.2 15.7 16.1 9.8 8.7
250 226 24 13 47 38
(88.3%) (88.6%) (85.7%) (27.1%) (40.9%) (17.6%)
QFT-GIT IND
1
y
Positive (%) 245 221 24 14 48 34
(86.6%) (86.7%) (85.7%) (29.2%) (41.7%) (15.7%)
IND 3 2
Y. Liu et al. / Tuberculosis 95 (2015) 523e526 Table 2 Agreement between the TB-IGRA and QFT-GIT results (n ¼ 656). TB-IGRA results þ QFT-GIT results þ 325 20 Total 345
Total
16 295 311
341 315 656
Agreement (95% CI)
Kappa value (95% CI)
95% (93%e96%)
0.89 (0.86e0.92)
infection risk (HCWs, 40.9% of TB-IGRA and 41.7% of QFT-GIT), and lowest in the populations with normal MTB infection risk (non-TB lung disease and healthy individuals) (27.1% and 17.6% of TB-IGRA, 29.2% and 15.7% of QFT-GIT). The results indicates that TB-IGRA has the comparable sensitivity and ability to reflect the risk levels of MTB infection in different populations with QFT-GIT, in another word, the two assays, TB-IGRA and QFT-GIT, have the same sensitivity and specificity. For further comparison of TB-IGRA and QFT-GIT, the result agreement and response correlation between the two tests were analyzed by SPSS 11.0 (SPSS Inc., Chicago, IL, USA) in 656 subjects without indeterminate results on any the two assays. Concordance between the tests was measured using the Kappa index (k value > 0.75, excellent agreement; 0.75 k 0.4, fair to good agreement; k < 0.4, poor agreement), and the correlation between the TB-IGRA response and the QFT-GIT response was analyzed nonparametrically by Spearman's correlation. All of the significance tests were two sided, and p < 0.05 was considered to be statistically significant. In the 656 subjects, both TB-IGRA and QFT-GIT were positive in 325 and negative in 295 subjects, and the observed agreement between the tests was 95%, with excellent concordance (k ¼ 0.89, 95% CI: 0.86e0.92) (Table 2). When investigating different populations, the agreement was 94%, 100%, 84%, 92% and 97% in PTBs, EPTBs, non-TB lung diseases, HCWs and healthy individuals, respectively (data not shown). The response correlation between the two assays was analyzed and described in Table 3. The IFN-g concentrations (T-N) of TB-IGRA were highly correlated to those of QFT-GIT in all subjects with r ¼ 0.85 (95% CI: 0.82e0.87; p < 0.001). The correlation coefficient in both positive subjects was 0.71 (95% CI: 0.65e0.76; p < 0.001), which was higher than that in both negative subjects (r ¼ 0.22, 95% CI: 0.82e0.87; p < 0.001). Although the correlation coefficient decreased from 0.81 in active TB patients to 0.47 in healthy individuals with the decrease of their positive rates, the responses of TB-IGRA were correlated significantly with those of QFT-GIT in all study populations with different risks of MTB infection (all p < 0.0001) (Table 3). These results indicate that there are excellent concordance between the results of TB-IGRA and QFT-GIT and good correlation between the responses of the two tests. In principle, there is no
Table 3 Correlations between the TB-IGRA responses and the QFT-GIT responses. Population (n)
Spearman r (95% CI)
p value
Overall subjects (656) Both positive (325) Both negative (295) Active TB patients (280) PTBs (252) EPTBs (28) Non-TB lung disease (45) HCWs (115) Healthy individuals (216)
0.85 0.71 0.22 0.81 0.81 0.81 0.67 0.72 0.47
<0.0001 <0.0001 0.0002 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
(0.82e0.87) (0.65e0.76) (0.10e0.33) (0.76e0.85) (0.76e0.85) (0.62e0.91) (0.46e0.81) (0.62e0.80) (0.35e0.57)
525
significant difference between the two tests. Both assays employ ELISA to measure the concentration of IFN-g released from lymphocytes in heparinized whole blood stimulated with MTB antigens compared with controls. The major difference between the two assays is the antigens used to stimulate the lymphocytes. TBIGRA uses a recombinant fusion protein of the region of difference-1 (RD1) antigens CFP-10 and ESTA-6 as the stimulus, whereas QFT-GIT uses pooled peptides from CFP-10 and ESAT-6, as well as an additional antigen TB7.7 (Rv2654c, which is not an RD1 antigen). But, previous studies have demonstrated that T-cell responses to recombinant antigens CFT-10 and ESTA-6 are similar to those to corresponding mixtures of synthetic overlapping peptides in active TB patients [16] and healthy subjects with latent TB infection [17]. In this study, most of discrepant results occurred in the individuals with low IFN-g response (see Supplementary Figure 1), which may be due to the decreased quantitative accuracy of IFN-g ELISA assays around their lower limit of detection rather than the difference of antigens used by TB-IGRA and QFT-GIT. 3. Conclusion In summary, TB-IGRA, a recently licensed Chinese domestic TB IFN-g release assay, has similar clinical performance to QFT-GIT in China. Due to this assay's low cost and high quality, it may be a useful tool in active TB diagnosis and MTB infection detection in low-income countries, where TB is endemic. Financial support: This work was supported by a grant from the Institute Reconstruction Fund (2011FU125Z04), Xiamen City Municipal platform fund on viral biotherapeutics (3502Z20131001) and National Science and Technology Major Project (2008ZX10003). Conflict of interest: We declare that we have no financial or personal relationships with other people or organizations that can inappropriately influence our work. There is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript. Ethical approval: This study received ethical approval from the Ethics Committee of the School of Public Health, Xiamen University. Written Informed consent was obtained from each participant. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.tube.2015.05.007. References [1] WHO. Global tuberculosis report 2013. 2013. [2] O'Garra A, Redford PS, McNab FW, Bloom CI, Wilkinson RJ, Berry MP. The immune response in tuberculosis. Annu Rev Immunol 2013;31:475e527. PubMed PMID: 23516984. [3] Dye C, Williams BG. Eliminating human tuberculosis in the twenty-first century. Journal of the Royal Society. Interface/R Soc 2008 Jun 6;5(23):653e62. PubMed PMID: 17690054. Pubmed Central PMCID: 3226985. [4] Gordon SV, Brosch R, Billault A, Garnier T, Eiglmeier K, Cole ST. Identification of variable regions in the genomes of tubercle bacilli using bacterial artificial chromosome arrays. Mol Microbiol 1999;32(3):643e55. [5] Denkinger C, Dheda K, Pai M. Guidelines on interferon-g release assays for tuberculosis infection: concordance, discordance or confusion? Clin Microbiol Infect 2011;17(6):806e14. [6] Metcalfe JZ, Everett CK, Steingart KR, Cattamanchi A, Huang L, Hopewell PC, et al. Interferon-gamma release assays for active pulmonary tuberculosis diagnosis in adults in low- and middle-income countries: systematic review
526
[7] [8]
[9]
[10]
[11]
[12]
Y. Liu et al. / Tuberculosis 95 (2015) 523e526 and meta-analysis. J Infect Dis 2011 Nov 15;204(Suppl. 4):S1120e9. PubMed PMID: 21996694. Pubmed Central PMCID: 3192542. WHO. Use of tuberculosis interferon-gamma release assays (IGRAs) in lowand middle income countries. 2011. ISBN 978 92 4 150267 2. Steffen RE, Caetano R, Pinto M, Chaves D, Ferrari R, Bastos M, et al. Costeffectiveness of Quantiferon(R)-TB Gold-in-Tube versus tuberculin skin testing for contact screening and treatment of latent tuberculosis infection in Brazil. PLoS One 2013;8(4):e59546. PubMed PMID: 23593145. Pubmed Central PMCID: 3617186. Song Q, Guo H, Zhong H, Liu Z, Chen X, Wang C, et al. Evaluation of a new interferon-gamma release assay and comparison to tuberculin skin test during a tuberculosis outbreak. Int J Infect Dis IJID Off Publ Int Soc Infect Dis 2012 Jul;16(7):e522e6. PubMed PMID: 22542006. Wolf T, Goetsch U, Oremek G, Bickel M, Khaykin P, Haberl A, et al. Tuberculosis skin test, but not interferon-gamma-releasing assays is affected by BCG vaccination in HIV patients. J Infect 2013 Apr;66(4):376e80. PubMed PMID: 23159357. Munoz L, Santin M. Interferon-gamma release assays versus tuberculin skin test for targeting people for tuberculosis preventive treatment: an evidencebased review. J Infect 2013 Apr;66(4):381e7. PubMed PMID: 23298892. Legesse M, Ameni G, Mamo G, Medhin G, Bjune G, Abebe F. Community-based cross-sectional survey of latent tuberculosis infection in Afar pastoralists, Ethiopia, using QuantiFERON-TB Gold In-Tube and tuberculin skin test. BMC
[13]
[14] [15] [16]
[17]
Infect Dis 2011;11:89. PubMed PMID: 21477326. Pubmed Central PMCID: 3080306. Mekaini LA, Jabri ON, Narchi H, Kamal SM, Mabrook A, Kuwaiti MM, et al. The use of an interferon-gamma release assay to screen for pediatric latent tuberculosis infection in the eastern region of the Emirate of Abu Dhabi. Int J Infect Dis IJID Off Publ Int Soc Infect Dis 2014 Jun;23:4e7. PubMed PMID: 24657274. Huebner RE, Schein MF, Bass Jr JB. The tuberculin skin test. Clin Infect Dis Off Publ Infect Dis Soc Am 1993 Dec;17(6):968e75. PubMed PMID: 8110954. Cellestis. QuantiFERON®-TB gold (In-Tube method) [package insert]. Carnegie, VIC, Australia: Cellestis; 2006. Arend SM, Geluk A, van Meijgaarden KE, van Dissel JT, Theisen M, Andersen P, et al. Antigenic equivalence of human T-cell responses to Mycobacterium tuberculosis-specific RD1-encoded protein antigens ESAT-6 and culture filtrate protein 10 and to mixtures of synthetic peptides. Infect Immun 2000 Jun;68(6):3314e21. PubMed PMID: 10816479. Pubmed Central PMCID: 97589. Arend SM, Engelhard AC, Groot G, de Boer K, Andersen P, Ottenhoff TH, et al. Tuberculin skin testing compared with T-cell responses to Mycobacterium tuberculosis-specific and nonspecific antigens for detection of latent infection in persons with recent tuberculosis contact. Clin Diagn Lab Immunol 2001 Nov;8(6):1089e96. PubMed PMID: 11687445. Pubmed Central PMCID: 96231.