ORIGINAL ARTICLE 10.1111/j.1469-0691.2006.01391.x Accuracy of an immune diagnostic assay based on RD1 selected epitopes for active tuberculosis in a clinical setting: a pilot study D. Goletti 1,2, S. Carrara 2, D. Vincenti 2, C. Saltini 3, E. Busi Rizzi 4, V. Schininà 4, G. Ippolito 5, M. Amicosante 3 and E. Girardi 5 1 Second Division of Health Department, 2 Translational Research Unit, 4 Radiology Department and 5 Epidemiology Department, National Institute for Infectious Diseases Lazzaro Spallanzani, Rome, and 3 Department of Internal Medicine, University of Tor Vergata, Rome, Italy ABSTRACT A previous case-control study reported that an in-vitro interferon (IFN)-c response to early secreted antigenic target (ESAT)-6 selected peptides was associated with active tuberculosis (A-TB). The objective of the present pilot study was to evaluate the diagnostic accuracy of this assay for TB disease in a clinical setting. An IFN-c ELISPOT assay was performed on samples from patients with suspected A-TB using two peptides selected from ESAT-6 protein and three peptides selected from culture filtrate 10 (CFP-10) proteins. The results were compared with those obtained by two commercially available assays approved for diagnosis of TB infection (T SPOT-TB and QuantiFERON-TB Gold) which use ESAT- 6 CFP-10 (RD1) overlapping peptides. Sensitivity to the RD1 selected peptides was 70% (positive for 16 of 23 patients with microbiologically diagnosed A-TB) and specificity was 91% (positive for three of 32 controls). In contrast, the sensitivity and specificity were 91% and 59%, respectively, for T SPOT-TB, and were 83% and 59%, respectively, for QuantiFERON-TB Gold. The RD1 selected peptides assay had the highest diagnostic odds ratio for A-TB. Thus, the results suggest that an assay based on RD1 selected peptides has a higher diagnostic accuracy for A-TB in a clinical setting compared with commercially available assays based on RD1 overlapping peptides. Keywords Diagnosis, ELISPOT, IFN-c, immunodiagnosis, RD1 selected peptides, tuberculosis Original Submission: 13 July 2005; Revised Submission: 15 September 2005; Accepted: 18 October 2005 Clin Microbiol Infect 2006; 12: 544 550 INTRODUCTION Epidemiological surveys suggest that one-third of the world s population is infected with Mycobacterium tuberculosis [1]. Primary infection leads to active tuberculosis (A-TB) in a minority (c. 10%) of infected individuals [2], usually within 2 years. In the remaining cases, the immune system contains the infection, and the individual is non-infectious and symptom-free [1]. This clinical latency can persist throughout an individual s lifetime. However, in some circumstances, the host immune response is perturbed and latent tuberculosis infection (LTBI) may develop into Corresponding author and reprint requests: D. Goletti, Laboratorio di Collegamento Tra Ricerca di Base e Clinica Padiglione Del Vecchio, Istituto Nazionale per le Malattie Infettive Lazzaro Spallanzani, I.R.C.C.S., Roma 00149, Italy E-mail: d.goletti@tiscali.it. post-primary A-TB. This process can occur, for example, following infection with human immunodeficiency virus (HIV), malnutrition, use of steroids or other immunosuppressive medications, or because of advanced age [3]. Development of new diagnostic tools would improve the control of tuberculosis (TB) by improving diagnosis of A-TB and by allowing a more accurate identification of LTBI [4]. Until recently, the tuberculin skin test (TST) has been the only tool available for diagnosing TB infection. However, this test is limited by its low sensitivity (75 90%) in diagnosing A-TB [5 7], and its low specificity, caused by the fact that the purified protein derivative (PPD) used for skin testing contains > 200 antigens that are shared widely among environmental mycobacteria [5 7] and the Mycobacterium bovis BCG strain used for vaccination. Lastly, TST does not allow discrimination between A-TB and LTBI [5]. Ó 2006 Copyright by the European Society of Clinical Microbiology and Infectious Diseases
Goletti et al. Immunological diagnosis of active TB 545 In an effort to develop more accurate tools for immunological diagnosis of TB, several studies have been performed to investigate early secreted antigenic target 6 (ESAT-6) and culture filtrate 10 (CFP-10) proteins [8 10]. These proteins, encoded by genes located within the region of difference 1 (RD1) of the M. tuberculosis genome, are much more specific for M. tuberculosis than is PPD, as they are not shared with BCG substrains or with most environmental mycobacteria (with the exception of Mycobacterium kansasii, Mycobacterium szulgai, Mycobacterium flavescens and Mycobacterium marinum). Peripheral blood mononuclear cells (PBMCs) of patients with TB [11,12], or of household contacts of TB patients [13], release interferon (IFN)-c when exposed in vitro to either ESAT-6 or CFP-10 intact proteins, or to overlapping peptides spanning the length of these antigens. The above studies resulted in the development of two commercially available tests (QuantiFER- ON-TB Gold and T SPOT-TB) that have been approved for the diagnosis of TB infection [10,14]. Both tests are based on ESAT-6 and CFP-10 overlapping peptides spanning the whole proteins, and employ ELISA or ELISPOT techniques. Although these assays provide an accurate diagnosis of M. tuberculosis infection, they do not discriminate between A-TB and LTBI. Thus, a positive response to these tests requires further clinical investigation to rule out A-TB. In this context, an IFN-c ELISPOT assay was designed, based on multi-epitopic peptides from the ESAT-6 protein, selected by quantitative implemented analysis of HLA peptide-binding motifs. A positive response with this assay was observed for most A-TB patients, while no response was observed with controls, even among TST-positive non BCG-vaccinated individuals [15]. Comparable observations were made with a whole-blood assay in which new CFP-10 selected peptides were tested [16]. Moreover, after therapy for 3 months, the IFNc response became undetectable in those patients who responded to treatment [17]. These data suggested that the response to ESAT-6 and CFP-10 selected peptides may be associated specifically with A-TB, but were obtained in case-control studies which tend to overestimate diagnostic accuracy [18]. The objective of the present pilot study was to evaluate whether an in-vitro assay based on RD1 selected peptides is potentially useful for the diagnosis of A-TB in a clinical setting. Thus, consecutive patients enrolled prospectively with suspicion of pulmonary TB were studied. The assay was performed using ESAT-6 CFP-10 (RD1) intact proteins and selected peptides [15 17], and the results obtained were compared with those obtained using two commercially available tests based on RD1 overlapping peptides that have been approved for in-vitro immunological diagnosis of TB infection. MATERIALS AND METHODS Patient population and study design Patients admitted to the infectious and respiratory diseases wards at the National Institute for Infectious Diseases (INMI) L. Spallanzani (Rome, Italy) between April 2004 and November 2004 were included in the study if there was a clinical suspicion of TB according to the institutional protocol, i.e., a chest radiograph suggestive of A-TB, with upper lobe infiltrates, cavitary lesions, hilar lymphadenopathy and miliary pattern, and or other signs and symptoms suggestive of A-TB, including fever, persistent cough, haemoptysis or weight loss [19]. Other criteria for inclusion were a negative HIV test and the absence of therapy with immunosuppressive drugs. Patients included in the study underwent clinical and microbiological examinations to confirm or exclude the diagnosis of A-TB. In brief, sequential respiratory samples (three expectorated or two induced sputa) during the first 7 days following hospital admission were collected. An acid-fast bacilli (AFB) smear examination (Ziehl-Neelsen) and culture, both on Lowenstein-Jensen medium and in a BACTEC 460 (Becton Dickinson, Sparks, MD, USA) with Middlebrooks 7H10 7H11 medium, were performed for each specimen. A chest radiograph and a TST, administered by the Mantoux method (5 Tuberculin Units), were performed for each patient. Individuals with a TST induration 10 mm in size were classified as TSTpositive [5 7]. Patients were classified as having A-TB if the diagnosis was confirmed by isolation of M. tuberculosis from biological specimens. For controls, patients were chosen who were admitted with suspected A-TB, but who subsequently yielded sputa negative by AFB smear and culture for M. tuberculosis, with either a resolution of clinical symptoms and radiographic abnormalities following antibiotic therapy not involving drugs active against M. tuberculosis, or who presented with an alternative diagnosis (e.g., lung cancer). Within 7 days of admission, and before starting anti-tb therapy, an EDTA blood sample was obtained from all enrolled individuals. An ELISPOT assay, based on RD1 selected peptides and intact proteins, was performed in parallel with the commercially available assays (QuantiFER- ON-TB Gold, Cellestis Ltd, Carnegie, Victoria, Australia; T SPOT-TB, Oxford Immunotec, Oxford, UK). Clinicians were unaware of the results of the in-vitro assays, and laboratory personnel were unaware of the clinical status of the patient. The study was approved by the local Ethics Committee, and all patients enrolled gave written informed consent.
546 Clinical Microbiology and Infection, Volume 12 Number 6, June 2006 Peptide selection and ELISPOT assay The selection of HLA-class II restricted epitopes of ESAT-6 and CFP-10 M. tuberculosis proteins was performed by quantitative implemented analysis of HLA peptide-binding motifs, as described previously for ESAT-6 [15 17]. Peptides were synthesised as free amino-acid termini using Fmoc chemistry (ABI, Bergamo, Italy). Lyophilised peptides were diluted in dimethyl sulphoxide (DMSO) at stock concentrations for each peptide of 10 mg ml, and were stored at )80 C. RD1 peptides used in this study are listed in Table 1. PBMCs were separated and stimulated as described previously [15 17]. In brief, cell cultures were treated in duplicate with: (i) a pool of the two ESAT-6 peptides (each at 25 mg L); (ii) a pool of the three CFP-10 peptides (each at 2 mg L); (iii) phytohaemagglutinin (PHA) (Sigma, St Louis, MO, USA) at 5mg L; and (iv) ESAT-6 and CFP-10 intact proteins (Lionex, Braunschweig, Germany) at 1 mg L. Scoring of results Results were scored as negative if the average of the raw number of spot-forming cells (SFCs) of cultures stimulated with PHA, ESAT-6 and CFP-10 (RD1) intact proteins and peptides was < 10. Results were scored as positive for RD1 peptides if the average number of SFCs was two-fold higher for the ESAT-6 peptide pool, or four-fold higher for the CFP-10 peptide pool, compared with the average number of SFCs for the corresponding controls. Results were scored as positive for PHA and RD1 intact proteins if the average number of SFCs in cultures stimulated with these antigens was three-fold higher than the average number of SFCs for the corresponding controls. A cut-off value was defined by constructing a receiver operator characteristic (ROC) curve to maximise the diagnostic accuracy for active tuberculosis. In this analysis, performed with LABROC-1 software (University of Chicago, Chicago, IL, USA), 34 SFCs million PBMCs for RD1 peptide pools, and 36 SFCs million PBMCs for intact proteins, were identified as the values maximising the sum of the specificity and sensitivity for A-TB. A concordance of 100% was found between the two scoring methods; therefore, the cut-off value was used to score the results. Results are presented as the number of SFCs million PBMCs, after subtraction of the appropriate control according to the criteria described above [15 17]. Statistical analysis The mean ± SD was calculated. Differences among groups were evaluated with the Mann Whitney U-test, while the McNemar test was used to compare the results of the different assays. Analysis was carried out with SPSS v.11 for Windows Table 2. Epidemiological and demographic characteristics of patients enrolled with a suspicion of active pulmonary tuberculosis (TB) (SPSS Inc., Chicago, IL, USA). To provide a global measure of the accuracy of immunological assays for the diagnosis of A-TB, the diagnostic odds ratio was calculated for each assay, defined as the odds of a positive test among patients with A-TB divided by the odds of a positive test among patients without A-TB. RESULTS Active TB patients Control patients TST-positive TST-negative Total (n) 23 18 14 Mean age ± SE (years) 33 ± 2 44 ± 3 38 ± 4 Gender Female (n) 8 6 7 Male (n) 15 12 7 Country of birth Western Europe (n) 5 16 9 Eastern Europe (n) 10 2 2 Africa (n) 4 2 Asia (n) 2 1 South America (n) 2 CD4 cell count ± SE 971 ± 149 993 ± 93 853 ± 314 BCG-vaccinated (n) 18 4 4 Diseases other than TB Bacterial pneumonia (n) 7 8 Bronchitis (n) 4 4 Fever of unknown origin (n) 2 2 Bacterial pleuritis (n) 1 Pulmonary carcinoma (n) 3 Pseudo-tumour (n) 1 TST, tuberculin skin test; SE, standard error; BCG, Bacillus Calmette et Guerin vaccine; CD, cluster of differentiation. Evaluation of response to RD1 selected peptides and intact proteins by ELISPOT assay Fifty-five patients were enrolled in the study (Table 2). TB diagnosis was confirmed microbiologically for 23 patients, and ruled out for 32 patients, 18 of whom were TST-positive and 14 of whom were TST-negative. When patient specimens were analysed by ELISPOT assay using the RD1 selected peptides, the number of IFN-c SFCs was significantly higher (131 ± 24) in patients with A-TB compared with controls (27 ± 13; p < 0.0001). This difference was statistically significant, independent of whether the controls Protein Position Sequence DR-serological specificities covered a ESAT-6 6 28 WNFAGIEAAASAIQGNVTSIHSL 1, 3, 4, 8, 11(5), 13(6), 52, 53 ESAT-6 66 78 NALQNLARTISEA 3, 8, 11(5), 13(6), 15(2), 52 CFP-10 18 31 FERISGDLKTQIDQ 3, 5, 11(5), 52 CFP-10 43 70 GQWRGAAGTAAQAAVVRFQEAANKQKQE 1, 3, 4, 7, 8, 11(5), 13(6), 15(2), 52 CFP-10 74 86 TNIRQAGVQYSRA 3, 4, 7, 8, 11(5), 12(5), 13(6), 15(2) Table 1. CD4 + T-cell epitopes in ESAT-6 and CFP-10 proteins selected to be broadly recognised in patients with active tuberculosis a Listed are the HLA-class II allele specificities to which the selected peptides are able to bind. CD, cluster of differentiation; ESAT, early secreted antigenic target; CFP, culture filtrate protein; HLA, human leukocyte antigen.
Goletti et al. Immunological diagnosis of active TB 547 were TST-positive (44 ± 23; p < 0.0001) or TSTnegative (6 ± 2; p < 0.0001) (Fig. 1A). The response to RD1 selected peptides was scored as positive for 16 (70%) of the 23 A-TB patients and for three (9%) of the 32 controls (3 18 TST-positive and 0 14 TST-negative). Thus, the sensitivity of this assay for A-TB in the population studied was 70%, and the specificity was 91% (Table 3). The RD1 intact proteins were used as an internal control. The number of IFN-c SFCs was significantly higher in patients with A-TB (181 ± 27) compared with controls (77 ± 18); this difference remained significant when compared with the TST-negative controls (13 ± 4; p < 0.0001), but not when compared with the TST-positive controls (114 ± 23; p 0.1) (Fig. 1B). A positive response to RD1 intact proteins was observed in 19 (83%) of the 23 A- TB patients and in 14 (44%) of the 32 controls. Thus, the sensitivity of the assay for A-TB was 83% and the specificity was 66% (Table 3). When considering patients with A-TB, the proportion of positive responses to RD1 intact proteins did not differ significantly from that of the RD1 selected peptides (p 0.37). In contrast, controls were significantly more likely to respond to RD1 intact proteins than to RD1 selected peptides (p 0.001), and the statistical significance was even greater (p 0.0001) if only TST-positive controls were considered. Comparison of IFN-c ELISPOT response to RD1 selected peptides and intact proteins with assays approved for the diagnosis of TB infection Fig. 1. IFN-c response to RD1 selected peptides in active tuberculosis (A-TB) patients and controls. Detection of IFN-c was by the ELISPOT assay. The highest response to (A) RD1 selected peptides, and (B) RD1 intact proteins in terms of specific interferon (IFN)-c-secreting T-cells is reported as spot-forming cells (SFCs) per million peripheral blood mononuclear cells (PBMCs) for each individual. Horizontal bars represent the mean of the SFCs for each group of patients. p values denote the difference among the responders in each group of patients. To substantiate the data, the results were compared with those obtained by T SPOT-TB and QuantiFERON-TB Gold (Table 3). A positive response to T SPOT-TB was observed with 21 (91%) of the 23 A-TB patients and with 13 (41%) of the 32 controls. A positive response to QuantiFERON-TB Gold was observed with 19 (83%) of the 23 A-TB patients and with 13 (41%) of the 32 controls. When the peptide responses were examined specifically in patients with A-TB, the proportion of positive responses to RD1 selected peptides was slightly lower compared with overlapping peptides, but the differences were not statistically significant (p 0.5 compared with both T SPOT-TB and QuantiFERON-TB Gold). However, when considering controls, the proportion of positive responses was significantly higher for T SPOT-TB and QuantiFERON-TB Gold than for the RD1 selected peptides assay (p 0.002 for both comparisons). This difference was still statistically significant when TST-positive controls were considered separately (p 0.002 and p 0.004 compared with T SPOT-TB and QuantiFERON-TB Gold, respectively). In contrast, no significant difference in the proportion of positive responses compared to overlapping peptides (T SPOT-TB and QuantiFERON-TB Gold) was found among either A-TB patients or controls when intact proteins were studied.
548 Clinical Microbiology and Infection, Volume 12 Number 6, June 2006 Table 3. Comparison of the different assays for immune diagnosis of tuberculosis (TB) Active TB patients Control patients TST-positive TST-negative Total Assay RD1 ELISPOT assay Selected peptides 16 23 (70) 3 18 (17) 0 14 (0) 3 32 (9) Intact proteins 19 23 a (83) 14 18 b (78) 0 14 (0) 14 32 b (44) RD1 commercially available assays T SPOT-TB 21 23 a (91) 13 18 b (72) 0 14 (0) 13 32 b (41) QuantiFERON-TB Gold 19 23 a (83) 13 18 b (72) 0 14 (0) 13 32 b (41) RD1, region of difference; TST, tuberculin skin test. a p > 0.05 compared with RD1 selected peptides assay. b p < 0.01 compared with RD1 selected peptides assay. Assay Sensitivity (95% Cl) Specificity (95% Cl) Diagnostic OR (95% Cl) RD1 ELISPOT assay Selected peptides 70% 91% 22 (47 87) (75 98) (5 97) Intact proteins 83% 56% 6 (61 95) (38 74) (2 22) RD1 commercially T SPOT-TB 91% 59% 15 available assays (72 99) (41 76) (3 77) QuantiFERON-TB Gold 83% 59% 7 (61 95) (41 76) (2 25) Table 4. Sensitivity, specificity and diagnostic odds ratio of the RD1 selected peptides assay for A-TB and control patients compared with the T SPOT-TB and QuantiFERON- TB Gold assays A-TB, active tuberculosis; RD1, region of difference. According to these data, the RD1 selected peptides ELISPOT assay had the highest diagnostic odds ratio for A-TB (Table 4). All of the assays performed similarly in identifying A-TB, but only the RD1 selected peptides assay may distinguish between A-TB and LTBI. DISCUSSION This pilot study assessed an assay based on RD1 selected peptides and intact proteins for the diagnosis of A-TB in a clinical setting. The ELISPOT technique was used to enumerate IFNc-secreting T-cells in response to these M. tuberculosis antigens. Use of selected peptides [15 17] gave a sensitivity for A-TB of 70% and a specificity of 91%. The accuracy of the response to RD1 selected peptides for active disease, as measured by the diagnostic odds ratio, was higher than for RD1 overlapping peptides used in tests approved for TB diagnosis (T SPOT-TB and QuantiFERON- TB Gold) [10,14]. Using an RD1 selected peptides immune assay, a slightly higher sensitivity among patients with A-TB has been reported [15 17]. However, the present study enrolled a consecutive series of patients recruited on the basis of suspected pulmonary A-TB, whereas the previous study, as well as others [20,21], used a case-control design, which can overestimate the diagnostic accuracy [18]. The immune assays based on overlapping peptides tended to have a higher sensitivity, although the difference with the RD1 selected peptides assay was not statistically significant. Importantly, four of the seven A-TB patients who did not give a positive response with the RD1 selected peptides were probably in a condition of M. tuberculosis-specific immune unresponsiveness, as they did not respond either to RD1 intact proteins or to the commercially available assays. However, it is important to note that patients with A-TB, as well as the controls, were all HIVnegative and were not receiving ongoing immunosuppressive therapy. An in-vitro response to the positive control, i.e., the mitogen PHA, was observed in all patients enrolled in the study, and similar CD4 cell counts were found among the groups of patients. Thus, the absence of response to RD1 antigens may also be explained as a consequence of the intrinsic limits of immune diagnostic assays, and is consistent with data published for patients with A-TB showing a 70% positive response to the RD1 ELISPOT assay [22]. An important feature of the RD1 selected peptides assay appears to be its specificity for A-TB, a characteristic that differentiates this test from other immune diagnostic assays designed to
Goletti et al. Immunological diagnosis of active TB 549 identify M. tuberculosis infection. The specificity of the RD1 selected peptides assay for active TB was 91%, compared with significantly lower figures for assays using RD1 intact proteins (56%) or overlapping peptides (59%). These data agree with a study based on the use of RD1 intact proteins in a clinical setting [23]. The results obtained using RD1 intact proteins were essentially in agreement with those obtained with the commercially available assays using RD1 overlapping peptides. Thus, the association of the assay results with A-TB is probably caused by the use of RD1 selected peptides rather than the particular assay format. At present, the reason for the differences in the response to the selected peptides and the overlapping peptides remains unclear, but could be related to the variable immunological status of the patients with A-TB and LTBI, or the selection of the cut-off value by ROC analysis (aimed at maximising the accuracy of the diagnosis of A- TB). Regarding the immune status of the patients, it is thought that metabolically active and viable bacilli secrete proteins, e.g., ESAT-6 and CFP-10 [24]. Consequently, IFN-c-producing T-cells specific to the selected epitopes are also present at high frequency only during the active replication of mycobacteria. Previous reports have demonstrated that lymphocytes with immediate effector memory function circulate for a limited time until the antigen has been cleared [25]. Therefore, the frequency of T-cells specific to the single epitopes contained in the selected peptides might decrease dramatically during latent infection in the absence of bacterial replication and continuous exposure to M. tuberculosis-secreted antigen. Thus, in the context of a low antigenic load, as in A-TB after efficacious treatment or in LTBI, the limited number of effector cells specific to the selected peptides in the peripheral blood could fall below the cut-off level, while the number of T-cells specific to the overlapping peptides covering the whole proteins, which contain multiple epitopes, could remain sufficiently high to be detected. This observation can also be applied to the use of intact proteins. Moreover, the cut-off value used in this study was chosen, following ROC analysis, to maximise the accuracy of A-TB diagnosis, whereas the cut-off values for the commercially available assays were probably chosen to maximise the accuracy of detection of TB infection [20,21]. The present study has several limitations. First, this is a pilot study performed in a limited population; thus, the estimates for sensitivity and specificity may be imprecise. Second, the invitro positive response to RD1 antigens could be related to infection with mycobacteria other than M. tuberculosis, e.g., M. kansasii, M. szulgai, M. flavescens, M. marinum and M. leprae, because the ESAT-6 gene, or a homologue, is also present in these related organisms [26,27]. Therefore, it will be important to evaluate patients with mycobacterial diseases other than TB to confirm the specificity of this assay. Notwithstanding these limitations, the development of a test using RD1 selected peptides may provide a significant improvement in TB diagnosis. Identification and treatment of individuals with latent M. tuberculosis infection are important goals in the control of TB, particularly in industrialised countries [4]. Treatment of LTBI requires that active disease is ruled out. Thus, in the context of a screening programme, an assay based on RD1 selected peptides may help to rule out A- TB among individuals who yield positive results in either the TST or RD1 overlapping peptides assay. Ideally, a new testing strategy would combine an assay based on RD1 intact proteins or overlapping peptides, which would provide the advantage of a higher sensitivity for TB infection, with an assay based on RD1 selected peptides, which may provide a higher specificity for the detection of active disease. ACKNOWLEDGEMENTS The authors are grateful to the patients and nursing staff who took part in this study. We thank D. Horejsh, R. Nisini, G. M. Fimia and T. Alonzi for their critical review of the manuscript. This work was supported by a grant from the Italian Ministry of Health. DG, SC, DV, MA and EG have a patent pending that describes a T-cell assay based on RD1 selected peptides. REFERENCES 1. Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC. Consensu statement. Global burden of tuberculosis: estimated incidence, prevalence and mortality by country. JAMA 1999; 282: 677 686. 2. Styblo K. Recent advances in epidemiological research in tuberculosis. Adv Tuberc Res 1980; 20: 1 63. 3. Flynn JL, Chan J. Immunology of tuberculosis. Ann Rev Immunol 2001; 19: 93 129. 4. Geiter L. The elimination of tuberculosis in the United States. In: Geiter L, ed., Ending neglect. Washington, DC: National Academy Press, 2000; 86 121.
550 Clinical Microbiology and Infection, Volume 12 Number 6, June 2006 5. American Thoracic Society. Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 2000; 161: 1376 1395. 6. Fine PE, Bruce J, Ponnighaus JM, Nkhosa P, Harawa A, Vynnycky E. Tuberculin sensitivity: conversions and reversions in a rural African population. Intl J Tuberc Lung Dis 1999; 3: 962 975. 7. Huebner RE, Schein MF, Bass JB. The tuberculin skin test. Clin Infect Dis 1993; 17: 968 975. 8. Behr MA, Wilson MA, Gill WP et al. Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science 1999; 284: 1520 1523. 9. Sorensen AL, Nagai S, Houen G, Andersen P, Andersen AB. Purification and characterization of a low-molecularmass T-cell antigen secreted by Mycobacterium tuberculosis. Infect Immun 1995; 63: 1710 1717. 10. Pai M, Riley LW, Colford JM. Interferon-gamma assays in the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis 2004; 4: 761 776. 11. Ravn P, Demissie A, Eguale T et al. Human T cell responses to the ESAT-6 antigen from Mycobacterium tuberculosis. J Infect Dis 1999; 179: 637 645. 12. Ulrichs T, Munk ME, Mollenkopf H et al. Differential T cell responses to Mycobacterium tuberculosis ESAT-6 in tuberculosis patients and healthy donors. Eur J Immunol 1998; 28: 3949 3958. 13. Doherty TM, Demissie A, Olobo J et al. Immune responses to the Mycobacterium tuberculosis-specific antigen ESAT-6 signal subclinical infection among contacts of tuberculosis patients. J Clin Microbiol 2002; 40: 704 706. 14. FDA. Approval for the use of synthetic peptide antigens used in the Quantiferon-TB Gold. http://www.fda.gov/cdrh/pma/ pmadec04.html. P10033 S0006. Date last update: 14 April 2005. Data last accessed: 14 April 2005. 15. Vincenti D, Carrara S, De Mori P et al. Identification of ESAT-6 epitopes for the immunodiagnosis of active tuberculosis. Mol Med 2003; 19: 105 111. 16. Goletti D, Vincenti D, Carrara S et al. RD1 selected peptides for active tuberculosis diagnosis: comparison between IFN-gamma whole blood ELISA and ELISPOT assays. Clin Diagn Lab Immunol 2005; 12: 1311 1316. 17. Carrara S, Vincenti D, Petrosillo N, Amicosante M, Girardi E, Goletti D. Use of a T-cell-based assay for monitoring efficacy of anti-tuberculosis therapy. Clin Inf Dis 2004; 38: 754 756. 18. Bossuyt PM, Reitsma JB, Bruns DE et al. The STARD statement for reporting studies of diagnostic accuracy: explanation and elaboration. Ann Intern Med 2003; 138: W1 W12. 19. Palmieri F, Girardi E, Petrosillo N et al. Linee-guida per la gestione clinica della infezione malattia tubercolare HIVassociata. Giorn It Mal Inf 2002; 8: 7 22. 20. Lalvani A, Pathan AA, Durkan H et al. Enhanced contact tracing and spatial tracking of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Lancet 2001; 357: 2017 2021. 21. Mori T, Sakatani M, Yamagishi F et al. Specific detection of tuberculosis infection with an interferon-gamma-based assay using new antigens. AJRCCM 2004; 170: 59 64. 22. Nicol MP, Pienaar D, Wood K et al. Enzyme-linked immunospot assay responses to early secretory antigenic target 6, culture filtrate protein 10, and purified protein derivative among children with tuberculosis: implications for diagnosis and monitoring of therapy. Clin Inf Dis 2005; 40: 1301 1308. 23. Ravn P, Munk EM, Andersen AB et al. Prospective evaluation of a whole-blood test using Mycobacterium tuberculosisspecific antigens ESAT-6 and CFP-10 for diagnosis of active tuberculosis. Clin Diag Lab Immunol 2005; 12: 491 496. 24. Andersen AB, Brennan P. Proteins and antigens of Mycobacterium tuberculosis. In: Bloom B, ed. Tuberculosis. Washington, DC: ASM Press, 1994; 307 332. 25. Pathan AA, Wilkinson KA, Klenerman P et al. Direct ex vivo analysis of antigen-specific IFN-gamma-secreting CD4+ T cells in Mycobacterium tuberculosis infected individuals: association with clinical disease state and effect of treatment. J Immunol 2001; 167: 5217 5225. 26. Geluk A, van Meijgaarden KE, Franken KI et al. Identification and characterization of the ESAT-6 homologue of Mycobacterium leprae and T-cell cross-reactivity with Mycobacterium tuberculosis. Infect Immun 2002; 70: 2544 2548. 27. Harboe M, Oettinger T, Wiker HG, Rosenkrands I, Andersen P. Evidence for occurrence of the ESAT-6 protein in Mycobacterium tuberculosis and virulent Mycobacterium bovis and for its absence in Mycobacterium bovis BCG. Infect Immun 1996; 64: 16 22.