Measurement of ESAT6-induced IFNγ responses adjunct with CXCL9 increases the rate of diagnosis of active tuberculosis in an endemic region Academic research paper on "Clinical medicine"

ELSEVIER

Available at www.sciencedirect.com

SciVerse ScienceDirect

journal homepage: www.elsevier.com/locate/IJMYCO

Mycobacteriology

Measurement of ESAT6-induced IFNy responses adjunct with CXCL9 increases the rate of diagnosis of active tuberculosis in an endemic region

Zahra Hasan a'\ Nisar Rao b, Naseem Salahuddin c, Mussarat Ashraf a, Muniba Islam a, Bushra Jamil a

a Aga Khan University, Karachi, Pakistan b Ojha Institute of Chest Diseases, DUHS, Karachi, Pakistan c Indus Hospital, Karachi, Pakistan

ARTICLE INFO

ABSTRACT

Article history: Received 14 May 2013 Accepted 26 May 2013 Available online 26 June 2013

Keywords: TB diagnosis CXCL9

Interferon-gamma

Due to difficulties in direct diagnosis of Mycobacterium tuberculosis infection where site-specific specimens are not available, indirect methods of testing for infection are required. M. tuberculosis early secreted antigen target-6 (ESAT6) induced IFN-y responses are specific, but do not differentiate between latent and active TB. The use of adjunct biomarkers for TB diagnosis has been proposed, such as the chemokines: CXCL9, CXCL10 and CCL2.

ESAT6-induced IFN-y CXCL9, CXCL10 and CCL2 was measured in whole blood cell super-natants of patients with pulmonary tuberculosis (PTB, n = 36) and extrapulmonary TB (ETB, n = 31) and compared with healthy endemic controls (EC, n = 33).

ESAT6-induced IFN-y responses were positive in 32% of TB cases as compared with 15% of EC cases (p = 0.048). ESAT6-induced CXCL9 responses were positive in 42% of TB cases and 15% of EC cases (p = 0.006). ESAT6-induced-CXCL10 and -CCL2 responses did not discriminate between TB and EC groups. Measurement of IFN-y or CXCL9 together diagnosed TB (53%) cases and was significant as compared with EC (p = 0.014) cases. IFN-y and CXCL10 together did not increase the number of TB cases diagnosed.

Within TB groups, ESAT6-IFN-y/CXCL9-based detection increased to 53% in PTB (p = 0.031) and 54% in ETB (p = 0.021), with comparable diagnosis in less severe extrapulmonary TB (L-ETB, 55%) and severe disseminated extrapulmonary TB (D-ETB, 50%). Given that 47% of TB cases remained undetected, this study shown that ESAT6-induced IFNy and CXCL9 can support diagnosis, but must be supported by clinical correlation and other relevant investigations.

© 2013 Asian-African Society for Mycobacteriology. Published by Elsevier Ltd. All rights

reserved.

Introduction

Tuberculosis (TB) causes almost 2 million deaths annually [1]. Diagnosis of TB remains a challenge owing to the difficulty in

detection of Mycobacterium tuberculosis, especially in paucibac-illary disease, when there are limited numbers of bacilli in clinical specimens, or access to site-specific specimens is difficult, such as in cases of extrapulmonary TB (ETB). Indirect

* Corresponding author. Address: Department of Pathology and Microbiology, Aga Khan University Hospital, P.O. Box 3500, Stadium Road, Karachi 74800, Pakistan. Tel.: +92 21 34930051x4511; fax: +92 21 34934294.

E-mail address: zahra.hasan@aku.edu (Z. Hasan). 2212-5531/$ - see front matter © 2013 Asian-African Society for Mycobacteriology. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijmyco.2013.05.002

biomarkers for diagnosis of TB are required for improved methods of detection, and a number of immune biomarkers have been identified to be potentially useful as diagnostic, predictive and prognostic markers of disease [2]. Ideally, the biomarker should be able to differentiate between healthy individuals and those with active disease. The biomarkers should also be useful in an endemic setting where there are high transmission rates and increased exposure to environmental mycobacteria. Initially, it was thought that inter-feron-gamma (IFN-y) responses to M. tuberculosis specific antigens would be useful for predicting active TB in individuals. IFN-y responses to the region of difference (RD)-1, recombinant antigen early secreted and T cell activated (ESAT)-6 and culture filtrate protein (CFP)-10 have been shown to be effective in detection of infected individuals [3-5]. Commercial IFN-y response assays (IGRA) based on ESAT-6 and CFP-10 are more specific for detecting M. tuberculosis infection [4] and are not affected by the bacille Calmette-Guerin (BCG) vaccination of the host, however, they do not discriminate between latent and active infection [6]. Hence, IGRAs are not recommended for diagnosis of TB. Also, these assays are less effective in cases where T-cell responses may be compromised such as in immune-compromised individuals with human immune deficiency virus (HIV) infection [7], owing to increased exposure to environmental mycobacteria [8] and in pediatric TB [9].

In the absence of effective Th1 cell immunity, it may be useful to have biomarkers of innate immune responses. An effective granulomatous response is essential for the restriction of infection with M. tuberculosis, and it is the appropriate activation of leucocytes via their chemokine receptors that leads to appropriate migration and localization [10]. A number of chemokines are activated by IFN-y, such as CXCL9, CXCL10, CXCL11 and CXCL12 [11]. CXCL9 levels are increased in patients with TB [12]. CXCL9 has been shown to be a marker for antigen-specific IFN-y production and IFN-y secreting cells [13]. The IFN-y inducible protein 10 (IP10 or CXCL10) is a marker of T-helper-1 (Th1) cells. It directs the migration and stimulates the adhesion of activated T-cells and natural killer cells by binding CXCR3, a G-protein coupled receptor. CXCL10 plays an important role in recruiting activated T cells into sites with tissue inflammation [11]. Recent reports have indicated the utility of CXCL10 for diagnosis of TB in latent infection [14] and in children with TB [15]. CCL2, or monocyte chemoattractant protein, is the most potent activator of monocytes and is responsible for recruitment of leucocytes through the CCR2 and CCR5 receptors to the site of infection [11]. CCL2 is activated by TNFa (tumor necrosis factor alpha) and plays an important role in protection against tuberculosis in the murine model [16]. CCL2 is differentially expressed in pulmonary and extrapulmonary tuberculosis [17]. CCL2 has also been suggested to be a biomarker for TB [18]. Both CXCL10 and CCL2 have been suggested to be useful for discrimination between individuals with and without tuberculosis infections [18,19].

Tuberculosis predominantly manifests itself as an infection of the lung parenchyma, but extrapulmonary forms of the disease remain common and can comprise up to 35% of cases [20]. As cytokine and chemokine responses to Mycobacterium and to mycobacterial antigens, ESAT-6 and

CFP-10 differ in patients with pulmonary (PTB) or extrapulmonary (EPTB) disease [19]. The utility of CXCL9, CXCL10 and CCL2 has been addressed as additional biomarkers in conjunction with ESAT6-induced IFN-y responses to diagnose TB in a highly endemic, BCG-vaccination population.

Materials and methods

Ethics statement

The study was approved by the Ethical Review Committee of The Aga Khan University and DOW University of Healthy Sciences (DUHS), Karachi, Pakistan. All patients provided written informed consent for the collection of samples and subsequent analysis.

Subject selection

Seventy seven patients were recruited from the Aga Khan University Hospital and Medical College (AKUH), OJHA Institute of Chest Disease, Dow University of Health Sciences (DUHS) and Indus Hospital, Karachi. Patients were newly diagnosed or had taken less than 7 days of anti-tuberculous therapy (ATT). Patients with significant co-morbid conditions, including diabetes mellitus, chronic renal failure, and chronic liver disease, and patients on high-dose corticosteroid therapy were excluded to ensure relatively unmodulated immu-nological parameters. All patients were HIV negative.

Patients with pulmonary TB (PTB, n = 36) were diagnosed by clinical examination, chest X-ray, sputum acid fast bacillus (AFB) Ziehl-Neelsen staining, AFB culture and/or clinical response to treatment (along with fever, cough and weight loss). All PTB patients had findings consistent with active TB as evaluated by one of the consulting physicians. Twenty-two patients had a positive AFB smear upon microscopy. Fourteen patients had a negative AFB smear, but showed a favorable clinical response to ATT. Patients were diagnosed as having minimal, moderate or advanced pulmonary tuberculosis using a modified classification of the National Tuberculosis Association of the USA based on extent of lung tissue involvement [21]. Of the PTB patients, 23 had moderately advanced, while 13 had far advanced disease.

Diagnosis of TB lymphadenitis, spinal and abdominal TB was based on histopathological findings of chronic granulom-atous inflammation with caseous necrosis, AFB staining and culture and supportive radiological evidence on CT or MRI. Diagnosis of meningeal TB was based on CSF biochemical findings, supported by AFB culture and findings on contrast-enhanced CT and/or MRI. Pleural TB was diagnosed on the basis of pleural fluid biochemical findings, AFB culture, histo-pathological findings on pleural biopsy, and supportive radiological evidence on X-rays and/or contrast-enhanced CT scan.

Patients with extrapulmonary TB (EPTB) were divided into two groups according to the WHO case definitions of severity of TB disease based on bacillary load, extent of disease and anatomical site [22]. EPTB patients were subsequently classified into less severe (L-ETB) or severe disseminated (D-ETB) disease. Therefore, the 31 ETB patients comprised those which had limited ETB (L-ETB) comprising tuberculous

lymphadenitis (n = 19) and unilateral pleural effusions (n = 12). Patients with disseminated extrapulmonary TB (D-ETB) comprised those with spinal TB (N = 4), abdominal TB (n = 2), miliary disease (N = 2), bilateral pleural effusions (n = 1) or tuberculous meningitis (n = 1).

BCG-vaccinated asymptomatic healthy volunteers who were staff at AKU with no known exposure to TB were used as endemic controls (ECs). All volunteers had a normal chest X-ray. All healthy endemic controls were BCG vaccinated.

Reagents

Recombinant antigen ESAT-6 from M. tuberculosis H37Rv was provided through the TB Vaccine Testing and Research Materials Contract, NIH, NIAID NO1-A1-40091, awarded to Colorado State University, USA.

Whole blood assay

Whole blood cultures were set up as described previously [23]. Cells were stimulated with recombinant antigens ESAT-6 (5 ig/ml) and cultured for 5 days and supernatants collected at 2- and 5-day post-stimulation for cytokine and chemokine measurements. All samples were set up in replicates. Samples were centrifuged to collect any cellular debris, aliquoted and stored at -70 °C until tested.

ELISA for IFN-y, CXCL9, CXCL10 and CCL2

IFN-y was detected in 5-day cellular supernatants by using standards and ELISA reagents obtained from Endogen (Rockford, IL, USA). IFN-y was measured using a sandwich ELISA technique according to the manufacturer's instructions and as reported previously [17]. Recombinant human cytokine was used to obtain a dose response curve with a range of detection from 3.9-1000 pg/ml. CXCL9, CXCL10 and CCL2 were measured in 2-day supernatants. Standards and monoclonal antibody pairs for capture and detection were obtained from R&D Systems (Abingdon, UK). All measurements were carried out according to the manufacturer's recommendations and as described previously [17]. Recombinant human cytokine was used to obtain a dose response curve with a range of detection

from 6.25-500 pg/ml for CXCL9, 4.4-1000 pg/ml for CXCL10 and 6.25-1000 pg/ml for CCL2.

Statistical analysis

Non-parametric statistical analysis was performed using Kruskal-Wallis and Mann-Whitney U tests, and Chi-Square test analyses were run using the Statistical Package for Social Sciences software (SPSS Version 17.0). Cytokine cut-off with sensitivity and specificity of the assays were determined using MEDCALC software version 11.4.

Results

Study subjects

The hematological parameters of study subjects are shown in Table 1. Patients comprised those with pulmonary TB (PTB) and extrapulmonary TB (EPTB). The age of patients and healthy endemic controls (EC) was comparable and there were no gender differences between patients with pulmonary or extrapulmonary TB.

ESAT6-induced IFN-y, CXCL9, CXCL10 and CCL2 in TB patients and ECs

ESAT6-induced cytokine responses in ex uiuo whole blood cultures were measured in patients with TB and in healthy endemic controls. ESAT6-induced IFN-y responses were found to be positive in 32% of TB patients and 15% in EC (p = 0.048). Measurement of ESAT6-induced chemokines illustrated that in TB patients 42% had positive CXCL9 responses, 8% were positive using CXCL10 and 23% had a positive CCL2 response. As shown in Table 2, these did not differ from those in healthy endemic controls. When IFN-y and CXCL9 were used together (in that either IFN-y or CXCL9) was positive, rates went up to 53% in TB patients and 27% in EC. A Chi-squared test run between IFN-y/CXCL9 responders in TB as compared with EC, this difference was significantly different (p = 0.014) When IFN-y and CXCL10 were used together, positivity rates were 33% in TB and 27% in EC, p = 0.53. When IFN-y and CCL2 were used, positivity rates were 45% in TB and 52% in EC.

Table 1 - Hematological characteristics of tuberculosis patients and healthy endemic controls.

Group EC PTB L-ETB D-ETB p-Value

N 33 36 31 10

Age (years) 30 24 27 32.5 0.055

Male, Female 17, 16 19, 17 12, 19 4, 6

ESR (mm/h) 7 56 42 41 0

Hb (g/dL) 13.2 11.4 11.5 11.8 0

TLC (10e9/L) 7 9.2 8 7.7 0.017

Lymphocytes (10e9/L) 2.3 1.7 1.9 1.8 0.004

Monocytes (10e8/L) 4.6 7.2 5.5 5.8 0.057

Neutrophils (10e9/L) 4 6.4 5.2 5.2 0.005

Values depict median data for each group. EC, healthy endemic controls; PTB, pulmonary TB; extrapulmonary TB with limited disease (L-ETB) or severe disseminated disease (D-ETB). ESR, erythrocyte sedimentation rate; Hb, hemoglobin; TLC, total lymphocyte count. "Denotes significant difference (p < 0.05) as determined by Kruskal-Wallis analysis.

Table 2 - ESAT6-induced IFN-y in combination with chemokines to diagnose active TB.

Positive responders (%)a

IFNy p-Value CXCL9 p-Value CXCL10 p-Value CCL2 p-Value IFNy-CXCL9 p-Value IFNy-CXCL10 p-Value

EC 15 15 15 39 27 27

TB 32 0.048 42 0.006 8 ns 23 ns 53 0.014 33 ns

PTB 31 ns 42 0.015 3 0.068 28 ns 53 0.031 31 ns

ETB 34 0.063 44 0.008 12 ns 20 ns 54 0.022 37 ns

L-ETB 39 0.033 42 0.017 16 ns 23 ns 55 0.088 42 ns

D-ETB 20 ns 50 0.022 0 ns 10 0 50 ns 20 ns

EC, healthy endemic controls; TB, tuberculosis patients; PTB, pulmonary TB; ETB, extrapulmonary TB; L-ETB, less severe extrapulmonary TB; D-ETB, severe disseminated extrapulmonary TB.

Cut-off values were: IFNy > 0 pg/ml; CXCL9 29 pg/ml; CXCL10 23 pg/ml; CCL2 2256 pg/ml. a A positive cut-off was based on MEDCALC software analysis, TB vs EC reference for each cytokine.

As CCL2 did not appear to be useful for identifying individuals with active TB in the population, it was not included in further calculations.

Disease site-specific variation in the utility of ESAT6-induced IFN-y, CXCL9 and CXCL10 responses

The number of positive responders that belonged to individuals with disease at different sites - pulmonary and extrapulmonary TB - were further investigated. Patients were stratified into PTB and EPTB patients comprising those with limited (L-ETB) or severe disseminated (D-ETB) disease. ESAT6-induced IFN-y, CXCL9 and CXCL10 responses in the different individuals were compared with those of ECs.

ESAT6-induced IFN-y responses alone indicated 31% responders in PTB, 39% in L-ETB and 20% in D-ETB groups, Table 2. Using CXCL9 alone, 42% of PTB; 42% of L-ETB; and 50% of D-ETB patients were found to be positive. Using CXCL10 alone, 3% of PTB, 16% of L-ETB but none in the D-ETB group were positive. With IFN-y in combination with CXCL9, positive rates were 53% of PTB; 55% of L-ETB and 25% of D-ETB as compared with EC 27%. In the case of IFN-y /CXCL10, positive rates were 31% of PTB; 42% of L-ETB and 20% of D-ETB as compared with 27% in EC.

Overall, the highest number of positive ESAT6-induced IFN-y and adjunct CXCL9 and CXCL10 responses were in L-ETB, followed by PTB and D-ETB groups.

Discussion

This study presents the diagnostic utility of ESAT6-induced IFN-y and adjunct chemokine responses for detection of TB in a highly endemic region in a low-HIV population. This data shows that measurement of ESAT6-induced IFN-y with CXCL9 increases the rate of detection of active tuberculosis. It was observed that ESAT6-induced IFN-y and ESAT6-induced CXCL9 responses were both significantly greater in TB patients as compared with healthy endemic controls.

Previously, studies in pediatric TB have reported CXCL10 to be a better alternate marker for TB diagnosis [15]. Also, that while CXCL10 alone was less effective than the Quantiferon assay for detection of M. tuberculosis infection, measuring both CXCL10 and IFN-y increased the number of TB patients

detected [24]. It was not found that measuring CXCL10 increased the number of cases detected; however, this study's focus was on adults and results may differ from those studies of children, who are more likely to have a greater innate and less-effective adaptive immune response driven by T-cells against Mycobacterium as compared with adults.

In a high transmission region, individuals are exposed at increased rates to both environmental mycobacterial and M. tuberculosis which results in modulation of antigen driven T-cell IFN-y responses [25]. This data corresponds with Goletti et al. which show a reduced diagnostic power of CXCL10 in a setting with high TB endemicity owing to the increased CXCL10 positive responses in those who did not have active TB [26].

It was found that ESAT6-induced CCL2 did not discriminate between individuals with and without active TB and also that this chemokine did not increase the number of detected cases when used as an adjunct marker to IFN-y. These finding are in contrast to findings by Ruhwald et al., in a study which comprised 8 patients with pulmonary TB and 7 healthy endemic controls [18]. However, the strength of the present study was greater (33 controls and 67 TB patients) than that of the former and thus may have more relevant information for an endemic region.

Most notable amongst these results was that measurement of CXCL9 with IFN-y increased the rate of active TB detection in patients with both pulmonary and extrapulmo-nary TB. This is especially pertinent for cases of extrapulmo-nary TB which pose diagnostic dilemmas owing to difficulty in obtaining site-specific specimens for microbiological tests. In the cases of L-ETB, IFN-y alone was sufficient to diagnose active TB; however, in D-ETB cases, CXCL9 measurement with IFN-y increased the rates of active TB detected. This fits with previous results which have shown that ESAT6-induced CXCL9 responses are increased in patients with severe disseminated TB, while IFN-y responses are reduced in the group [19].

Conclusion

This study reports the utility of the combined measurement of IFN-y and CXCL9 for detection of TB. As early diagnosis of extrapulmonary TB remains a challenge, a minimally invasive

method of diagnosing TB using blood-based biomarkers could be very useful for diagnosis of active TB. However, given that only 50% of all active TB cases were detected, these markers should not be used in isolation and must be analyzed in conjunction with clinical correlation and additional relevant investigations.

Ethical consideration

The study received approval from the Ethical Review Committees of the Aga Khan University and DOW University of Health Sciences. Written informed consent was taken from all study subjects.

Author's contribution

Z.H. designed and implemented the study. N.R., N.S. and B.J. were responsible for patient recruitment. M.A. and M.I. were responsible for laboratory experiments and statistical analysis.

Conflict of interest

The authors declare there are no conflicts of interest.

Acknowledgments

Thanks for technical assistance to Maqboola Dojki. Thanks for help with patient recruitment to Kiran Iqbal and Dr. Erum, Indus Hospital. Thanks to Najeeha Talat for help with statistical analysis. This study received financial support from the Asia-Links Program grant, Swedish Research Council.

REFERENCES

[1] World Health Organisation, Global Tuberculosis Control, World Health Organisation, 2011.

[2] G. Walzl, K. Ronacher, W. Hanekom, T.J. Scriba, A. Zumla, Immunological biomarkers of tuberculosis, Nat. Rev. Immunol. 11 (2011) 343-354.

[3] T. Ulrichs, P. Anding, S. Porcelli, S.H.E. Kaufmann, M.E. Munk, Increased numbers of ESAT6 and purified protein derivative-specific gamma interferon-producing cells in subclinical and active tuberculosis infection, Infect. Immun. 68 (2000) 60736076.

[4] S.M. Arend, A.C.F. Engelhard, G. Groot, D.K. Boer, P. Andersen, T.H. Ottenhoff, 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. 8 (2001) 1089-1096.

[5] O. Parkash, B.P. Singh, M. Pai, Regions of differences encoded antigens as targets for immunodiagnosis of tuberculosis in humans, Scan. J. Immunol. 70 (2009) 345-357.

[6] M. Sester, G. Sotgiu, C. Lange, C. Giehl, E. Girardi, G.B. Migliori, et al, Interferon-g release assays for the diagnosis of active

tuberculosis: a systematic review and meta-analysis, Eur. Respir. J. (2010).

[7] M.G. Aabye, P. Ravn, G. Praygod, K. Jeremiah, A. Mugomela, M. Jepsen, et al, The impact of HIV infection and CD4 cell count on the performance of an interferon gamma release assay in patients with pulmonary tuberculosis, PLoS ONE 4 (2009) e4220.

[8] G.F. Black, P.E.M. Fine, D.K. Warndorff, S. Floyd, R.E. Weir, J.M. Blackwell, et al, Relationship between IFNy and skin test responsiveness to Mycobacterium tuberculosis PPD in healthy, non-BCG-vaccinated young adults in Northern Malawi, Int. J. Tuberc. Lung Dis. 5 (2001) 664-672.

[9] A.R. Bamford, A.M. Crook, J.E. Clark, Z. Nademi, G. Dixon, J.Y. Paton, et al, Comparison of interferon-gamma release assays and tuberculin skin test in predicting active tuberculosis (TB) in children in the UK:

a paediatric TB network study, Arch. Dis. Child. 95 (2010) 180-186.

[10] H.M.S. Algood, J. Chan, J.L. Flynn, Chemokines and tuberculosis, Cyt. Growth Fact. Rev. 14 (2003) 467-477.

[11] B. Moser, M. Wolf, A. Walz, P. Loetscher, Chemokines: multiple levels of leukocyte migration control, Trends Immunol. 25 (2004) 75-84.

[12] A.L. Alessandri, A.L. Souza, S.C. Oliveira, G.C. Macedo, M.M. Teixeira, A.L. Teixeira, Concentrations of CXCL8, CXCL9 and sTNFR1 in plasma of patients with pulmonary tuberculosis undergoing treatment, Inflamm. Res.

55 (2006) 528-533.

[13] C. Abramo, K. Meijgaarden, D. Garcia, K. Franken, M. Klein, A. Kolk, et al, Monokine induced by interferon gamma and IFN-gamma response to a fusion protein of Mycobacterium tuberculosis ESAT-6 and CFP-10 in Brazilian tuberculosis patients, Microbes Infect. 8 (2006) 45-51.

[14] M. Ruhwald, T. Bodmer, C. Maier, M. Jepsen, M.B. Haaland, J. Eugen-Olsen, et al, Evaluating the potential of IP-10 and MCP-2 as biomarkers for the diagnosis of tuberculosis, Eur. Respir. J. 32 (2008) 1607-1615.

[15] E. Whittaker, A. Gordon, B. Kampmann, Is IP-10 a better biomarker for active and latent tuberculosis in children than IFNy?, PLoS ONE 3 (2008) e3901

[16] W. Peters, H.M. Scott, H.F. Chambers, J.L. Flynn, I.F. Charo, J.D. Ernst, Chemokine receptor 2 serves an early and essential role in resistance to Mycobacterium tuberculosis, Proc. Natl. Acad. Sci. USA 98 (2001) 7958-7963.

[17] Z. Hasan, I. Zaidi, B. Jamil, M.A. Khan, A. Kanji, R. Hussain, Elevated ex uiuo monocyte chemotactic protein-1 (CCL2) in pulmonary as compared with extra-pulmonary tuberculosis, BMC Immunol. 6 (2005) 14.

[18] M. Ruhwald, M. Bjerregaard-Andersen, P. Rabna, J. EugenOlsen, P. Ravn, IP-10, MCP-1, MCP-2, MCP-3, and IL-1A hold promise as biomarkers for infection with M. tuberculosis in a whole blood based T cell assay, BMC Res. Notes 2 (2009).

[19] Z. Hasan, B. Jamil, M. Islam, M.S. Yusuf, M. Ashraf, J.A. Khan, et al, ESAT6-induced CXCL9 and IFNy can differentiate severity of tuberculosis, PLoS ONE 4 (2009) e5158.

[20] WHO, Country Profile Pakistan, World Health Organisation, 2008. Global Tuberculosis Control, pp. 133-136.

[21] J. Crofton, Clinical features of tuberculosis, Crofton and Douglas Respiratory Diseases, Blackwell Scientific, London, 1990, pp. 395-421.

[22] World Health Organisation, in: D. Maher, P. Chaulet, S. Spinaci, A. Harries (Eds.), Treatment of Tuberculosis: Guidelines for National Programs, vol. 3, World Health Organisation, Geneva, Switzerland, 2003, pp. 1-77.

[23] R. Hussain, A. Kaleem, F. Shahid, M. Dojki, B. Jamil, M. Mehmood, et al, Cytokine profiles using whole-blood assays can discriminate between tuberculosis patients and healthy endemic controls in a BCG-vaccinated population, J. Immunol. Methods 264 (2002) 95-108.

[24] J. Lighter, M. Riguad, M. Huie, C.H. Peng, H. Pollack, Chemokine IP-10: an adjunct marker for latent tuberculosis infection in children, Int. J. Tuberc. Lung Dis. 13 (2009) 731736.

[25] G.F. Black, R.E. Weir, S.D. Chaguluka, D. Warndorff, A.C. Crampin, L. Mwaungulu, et al, Gamma interferon responses induced by a panel of recombinant and purified

mycobacterial antigens in healthy, non-Mycobacterium bouis BCG-vaccinated Malawian young adults, Clin. Diagn. Lab. Immunol. 10 (2003) 602-611. [26] D. Goletti, R. Alamelu, B.S.A. Kabeer, C. Rodrigues, A. Sodha, S. Carrara, et al, Is IP-10 an accurate marker for detecting M. tuberculosis-specific response in HIV-infected persons?, PLoS ONE 5 (2010) e12577