Skip Navigation

Clinical, Immunological, and Epidemiological Importance of Antituberculosis T Cell Responses in HIV-Infected Africans

  1. Molebogeng X. Rangaka1,5,a,
  2. Lavanya Diwakar1,a,
  3. Ronnett Seldon1,
  4. Gilles van Cutsem5,
  5. Graeme A. Meintjes1,6,
  6. Chelsea Morroni2,
  7. Priscilla Mouton1,5,
  8. Muki S. Shey1,
  9. Gary Maartens1,3,4,
  10. Katalin A. Wilkinson1,7,a, and
  11. Robert J. Wilkinson1,4,6,7
  1. 1Institute of Infectious Diseases and Molecular Medicine, Faculty of Health Sciences, Cape Town
  2. 2School of Public Health, University of Cape Town, Cape Town
  3. 3Division of Pharmacology, University of Cape Town, Cape Town
  4. 4Department of Medicine, University of Cape Town, Cape Town
  5. 5Medicins Sans Frontieres, Khayelitsha site B, Khayelitsha
  6. 6G.F. Jooste Hospital, Manenberg, South Africa
  7. 7Division of Medicine, Imperial College London, United Kingdom
  1. Reprints or correspondence: Dr. Robert J. Wilkinson, Rm. S2.19.4 Institute of Infectious Diseases and Molecular Medicine, Wernher and Beit Bldng. South, Faculty of Health Sciences, University of Cape Town, Anzio Rd., Observatory 7925, South Africa (r.j.wilkinson{at}imperial.ac.uk).

  • ↵a M.X.R., L.D., and K.A.W. contributed equally to this article.

 
Next Section

Abstract

Background. Human immunodeficiency virus (HIV)-associated tuberculosis is a major cause of mortality in Africa. The assay of T cell interferon-γ released in response to antigens of greater specificity than purified protein derivative is a useful improvement over the Mantoux tuberculin skin test, but few studies have evaluated interferon-γ secretion in HIV-infected individuals.

Methods. Mycobacterium tuberculosis antigen-specific interferon-γ secretion was assessed by whole blood assay and enzyme-linked immunospot, which were compared with the Mantoux tuberculin skin test in HIV-infected and HIV-uninfected individuals without active tuberculosis and HIV-infected patients with pulmonary tuberculosis in Khayelitsha, South Africa.

Results. The skin test and whole blood assay responses to purified protein derivative in HIV-positive subjects were decreased, compared with responses in HIV-negative subjects (P < .001). By contrast, the responses to M. tuberculosis antigens (early secreted antigenic target 6, culture filtrate protein 10, TB10.3, and α-crystallin 2) were less affected, indicating a high prevalence of latent tuberculosis (∼80%) in both HIV-negative and HIV-positive subject groups. Whole blood assay responses did not differ between the HIV-positive subjects without tuberculosis and HIV-positive subjects with tuberculosis, but the enzyme-linked immunospot method response to early secreted antigenic target 6 and culture filtrate protein 10 was higher in the group of HIV-infected subjects with tuberculosis (P ⩽ .04), although this group had lower CD4+ cell counts. A ratio of the combined enzyme-linked immunospot method response divided by the CD4+ cell count of >1.0 had 88% sensitivity and 80% specificity for active pulmonary tuberculosis in HIV-infected individuals.

Conclusions. Interferon-γ release appears to be less impaired than skin testing by HIV coinfection. The novel potential to relate the enzyme-linked immunospot method and CD4+ cell count to assist diagnosis of active tuberculosis in patients with HIV infection is important and deserves further evaluation.

Tuberculosis (TB) is the major cause of death in HIV-infected individuals in Africa. Although it has only 11% of the world's population, Africa accounts for more than one-quarter of the global burden of TB, with an estimated 2.4 million TB cases and 540,000 TB deaths annually [1]. Although it is very well recognized that the degree of CD4+ T cell depletion relates to the risk of HIV-associated TB [2, 3], surprisingly few studies have examined in detail the T cell response to Mycobacterium tuberculosis antigens in HIV-infected individuals in regions of endemicity. Some early studies of antigens not restricted to the M. tuberculosis complex showed the cellular proliferative and IFN-γ response to various antigenic preparations to be markedly decreased both in HIV-infected patients with active TB and in HIV-infected patients without TB [4, 5]. By contrast, other analyses reported proliferative responses and found that the production of TNF was 3–10 times higher in patients than in control subjects [6]. IFN-γ responses to the M. tuberculosis-specific antigens culture filtrate protein 10 (CFP-10) and early secreted antigenic target 6 (ESAT-6), measured by enzyme-linked immunospot (ELISPOT) assay, also appear to be relatively unimpaired by HIV infection [7, 8]. ESAT-6 and CFP-10 are encoded by a genomic segment (RD1) absent from all bacille Calmette-Guérin strains [9,10–11], and ELISPOT responses to these antigens differentiate immune sensitization by bacille Calmette-Guérin from infection by pathogenic mycobacteria [12]. In addition, sensitivity for the detection of overt TB infection (i.e., detection of culture-positive patients) is 80%–90% [12,13–14]. In outbreak circumstances, a positive ELISPOT response more clearly correlates with the degree of exposure to the index case than does the tuberculin skin test (TST) [15, 16].

An alternative to ELISPOT analysis is the determination of the IFN-γ content of supernatants derived from stimulated whole blood. Evidence suggests that whole blood assays based on ESAT-6 and CFP-10 are also of increased specificity and sensitivity, compared with the TST [17]. However, no study has systematically evaluated the whole blood assay response to ESAT-6 and CFP-10 in the context of HIV infection. This is important, because the skin response to tuberculin PPD used in the TST is impaired by HIV [18]. Were it to be convincingly shown that in vitro tests of sensitization were less impaired than other tests by HIV, this might justify their wider implementation in resource-poor environments, despite their inherently greater complexity and expense. Indeed, a goal of the global Stop-TB partnership is to implement, by 2012, “a diagnostic toolbox to accurately identify people with latent TB infection and those at high risk of progression to disease” [19, p. 16]. In Africa alone, this plan envisages that 21 million people living with HIV infection and/or AIDS will be screened for latent TB and, of these, 2.4 million people will be offered isoniazid preventive therapy in the year 2015. This will be difficult to achieve without technological improvement.

In this study, we employed a research whole blood assay that involves dilution of the blood, followed by culture at 72 h after collection of the sample (as opposed to overnight) [20, 21]. In addition to assaying the response to ESAT-6 and CFP-10, we included PPD and 2 additional novel and immunodominant antigens. TB 10.3 is a member of the ESAT-6 family, but it is not restricted to pathogenic mycobacteria [22]; it has recently been implicated in protection against experimental bovine TB [23]. The antigen α-crystallin 2 (Acr2) is a member of the α-crystallin family of small heat shock proteins that we have previously found to be expressed—and, therefore, recognized by T cells—shortly after M. tuberculosis infection [24, 25]. Our first aim was to determine differences in M. tuberculosis antigen recognition between HIV-infected and uninfected individuals living in the same area. Second, what differences exist between HIV-infected individuals with and HIV-infected individuals without active pulmonary TB? Third, we were interested in determining whether differences in the pattern of antigen recognition between groups correlated with TB disease status and whether we could thereby gain insight into the potential diagnostic utility of these responses.

Previous SectionNext Section

Materials and Methods

Participants. This study was approved by the University of Cape Town (Cape Town, South Africa) Research Ethics Committee (reference numbers 443/2004 and 173/2005). Written consent for inclusion was provided by all participants. Recruitment was conducted at the Ubuntu HIV-TB clinic in the socioeconomically deprived periurban township of Khayelitsha (Cape Town, South Africa). There is an extremely high burden of both TB and HIV infection in this township, with an annual TB incidence of 1612 cases per 100,000 population and an antenatal HIV seroprevalence of 33% in 2005. HIV-seropositive persons with newly diagnosed pulmonary TB (the HIV-PTB group) were enrolled within 5 doses after starting anti-TB chemotherapy; all of these patients were culture-positive or sputum smear-positive for M. tuberculosis. Persons with a new diagnosis of HIV infection but without signs or symptoms of active TB (the HIV-nonTB group) and healthy persons found to be HIV seronegative (the HIV-negative group) were enrolled sequentially shortly after attending the same voluntary counselling and testing clinic for HIV infection. Thus, we were able to recruit similar groups that differed only by their HIV-infection status. Both HIV-nonTB and HIV-negative subjects were screened for active TB by a symptom score questionnaire that was based on validated approaches that demonstrate that chest radiography does not increase diagnostic yield [26, 27]. The presence of any 1 of the following signs or symptoms formed exclusion criteria and led to referral: cough, chest pain, weight loss, night sweats, fever, loss of appetite, lymphadenopathy, and generalized tiredness. Furthermore, any other suspicious finding on clinical examination also gave rise to an investigation according to national guidelines. Receipt of current antibiotic, antiviral, and steroid therapy were exclusions, as were pregnancy, age <18 years, and any acute illness other than TB. HIV-nonTB subjects who qualified for antiretroviral therapy under current guidelines (CD4+ cell count <200 cells/mm3 or World Health Organization clinical stage 4) were referred to the antiretroviral clinic.

Mantoux TST. For each subject, a dose of 2TU tuberculin PPD RT23 was injected intradermally into the volar aspect of the left forearm. Transverse induration was recorded at 48 h. Positivity was defined as an induration ⩾15 mm in diameter in the HIV-negative group and ⩾5 mm in diameter in the HIV-nonTB group and the HIV-PTB group. HIV-nonTB subjects who experienced Mantoux reactions ⩾5 mm in diameter were offered isoniazid preventive therapy according to South African national guidelines.

IFN-γ release assays. Laboratory workers were blind to the clinical status of participants. ELISPOT and whole blood assay analysis were performed as previously described [21, 28, 29]. Antigenic stimuli were endotoxin free and were assayed in duplicate wells. The concentrations used were 5 µg/mL of ESAT-6, CFP-10, and TB10.3 and 20 µg/mL of Acr2. PPD was used at 200 IU/mL (4.6 µg/mL). Control wells included phytohaemaggutin (5 µg/mL) and no antigenic stimulus. A 3-day whole blood assay was performed [20], in which we found a minimal further increase in IFN-γ production when extending cultures to 5 days. ELISPOT plates were read on an Immunospot Series 3B Analyzer (Cellular Technology), and plates were retained for visual inspection and confirmation in case of anomaly.

Analysis. Because the aim of the study was to determine the effect of HIV infection on the recognition of TB antigens and to further compare the response in HIV-infected persons with and HIV-infected persons without active TB, the HIV-nonTB subject group was used as a reference for comparisons, and no comparisons between the HIV-negative group and the HIV-PTB group were made. Nonparametric unpaired data were analyzed by the Mann-Whitney U test. Proportions were compared by χ2 test or by Fisher's exact test of probability if the contingency included a number <5. Correlation of nonparametrically distributed variables was performed by Spearman analysis. Receiver-operator characteristic analysis was performed using Prism 4 software (GraphPad) [30].

Previous SectionNext Section

Results

Characteristics of the study groups. A total of 123 people were enrolled, with 41 patients in each of the 3 groups: HIV-infected persons with newly diagnosed, untreated pulmonary TB (the HIV-PTB group), persons with newly diagnosed HIV infection but without signs or symptoms of active TB (the HIV-nonTB group), and healthy HIV-uninfected subjects (the HIV-negative group) attending the same voluntary counselling and testing facility.

The median CD4+ cell count was lower in the HIV-PTB group than in the HIV-nonTB group (167 cells/mm3 vs. 464 cells/mm3; P < .001), as was the median body mass index, calculated as weight in kilograms divided by the square of the height in meters (20.7 vs. 27.3; P < .001). The groups did not significantly differ with respect to age, sex, bacille Calmette-Guérin infection status, or contact with TB-infected individuals (table 1). The Mantoux TST was considered to have a positive result if the diameter of the induration was ⩾5 mm in the HIV-nonTB group and ⩾15 mm in the HIV-negative group. By this definition, there was no statistically significant difference in the proportion of subjects with a positive TST result in each group. However, when analyzed as a quantitative variable, the median response in the HIV-nonTB group (induration diameter, 8 mm) was less than that in the HIV-negative group (induration diameter, 19 mm; P < .001). Patients in the HIV-PTB group showed a markedly skewed distribution of TST responses, which tended to be either completely negative or strongly positive. Although the median value was greater in the HIV-PTB group than in the HIV-nonTB group, this difference was not statistically significant.

Figure 1
View larger version:
Figure 1

IFN-γ enzyme-linked immunospot (ELISPOT) response to antigens of Mycobacterium tuberculosis. A, Data are shown as spot-forming cells (SFC) per 106 PBMCs. Shaded symbols indicate HIV-infected individuals with pulmonary tuberculosis (HIV-PTB group), and unshaded symbols indicate HIV-infected individuals without tuberculosis (HIV-nonTB). Horizontal lines indicate median values. The HIV-PTB group had a significantly greater response to all 3 antigens, compared with the HIV-nonTB group (P ⩽ .04). B, In the HIV-PTB group, ELISPOT responses had no discernible relationship to CD4+ cell count and were as likely to be high in individuals with severe immune depletion as in individuals with more moderate depletion of CD4+ cells. C, ELISPOT frequencies for each antigen were divided by the corresponding total CD4+ cell count and were plotted against CD4+ cell count. When considering all antigens, the median value of this ratio in the HIV-PTB group (shaded symbols) was 18.6 times greater than that in the HIV-nonTB group (unshaded symbols) (P < .001, by Mann-Whitney U test).

Figure 2
View larger version:
Figure 2

Percentage of matched HIV-infected (HIV+) and HIV-uninfected (HIV-) subjects attending voluntary counselling and testing with positive whole blood assay (WBA) and Mantoux tuberculin skin test (TST) results. A total of 92% of HIV- subjects had a Mantoux tuberculin skin test response of >5 mm in diameter, and 69% had a response of >15 mm in diameter; however, only 61% of HIV+ subjects had responses >5 mm in diameter (P = .004), and only 42% had responses >15 mm in diameter (P = .025). By contrast, the percentage of subjects with positive whole blood assay results by early secreted antigenic target 6 or culture filtrate protein 10 reactivity did not differ between HIV+ and HIV- subjects.

Table 1
View larger version:
Table 1

Demographic and clinical characteristics of subjects enrolled in the study.

IFN-γ secretion in whole blood assays. The whole blood assay was performed for 115 of 123 participants. Five subjects (12.2%) in the HIV-PTB group and 3 subjects (7.3%) in the HIV-nonTB group had an insufficient sample. There was a nonsignificant trend towards lower IFN-γ secretion in response to the M. tuberculosis-specific antigens in the HIV-nonTB group, compared with the HIV-negative group, with the exception of TB10.3, for which the responses in the 2 groups were similar (table 2). IFN-γ secretion in response to PPD, however, was markedly decreased in the HIV-nonTB group, compared with the HIV-negative group (an 11.4-fold decrease; 913 µg/mL vs. 10,433 µg/mL; P < .001). When comparing the HIV-nonTB group to the HIV-PTB group, patients with active TB had a stronger response to Acr2, TB10.3, and PPD and a weaker response to ESAT-6 and CFP-10. However, responses were markedly heterogenous within the groups, and none of these differences were statistically significant (table 2).

Table 2
View larger version:
Table 2

IFN-γ responses to Mycobacterium tuberculosis antigens in whole blood assay.

IFN-γ ELISPOT analysis. A total of 31 subjects in the HIV-PTB group and 16 subjects in the HIV-nonTB group provided a sufficient amount of blood to isolate adequate numbers of PBMCs to also perform an ELISPOT assay using ESAT-6 and CFP-10. The median responses to ESAT-6 (353 spot-forming cells [interquartile range {IQR}, 99–977 spot-forming cells] vs. 120 spot-forming cells [IQR, 20–369 spot-forming cells]; P = .038), CFP-10 (376 spot-forming cells [IQR, 45–1092 spot-forming cells] vs. 20 spot-forming cells [IQR, 3–110 spot-forming cells]; P = .005), and PPD (279 spot-forming cells [IQR, 130–693 spot-forming cells] vs. 25 spot-forming cells [IQR, 3–92 spot-forming cells]; P = .001) were significantly higher in the HIV-PTB group, compared with the HIV-nonTB group (figure 1A).

Stratification of responses by CD4+ cell count. The conditions of both IFN-γ release assays favor the detection of M. tuberculosis-specific CD4+ cells and will, therefore, be influenced by the absolute CD4+ cell count. Therefore, we stratified the whole blood assay and ELISPOT results by CD4+ cell count. In both the HIV-nonTB group and the HIV-PTB group, the responses to PPD, ESAT-6, and CFP-10 decreased with decreasing CD4+ cell count. PPD-stimulated whole blood IFN-γ values significantly correlated with CD4+ cell count in the HIV-PTB group (r = 0.4436; P = .012); this correlation was not statistically significant for ESAT-6 and CFP-10. In contrast, a lower-level response to TB10.3 and Acr2 was preserved irrespective of CD4+ cell count; the response to Acr2 was, in fact, slightly higher in those patients in the HIV-PTB group who had the greatest CD4+ cell depletion.

The ELISPOT response to PPD also correlated with CD4+ cell count in the HIV-nonTB group (r = 0.5489; P = .034); otherwise, correlations were nonsignificant, although the responses to ESAT-6 and CFP-10 again decreased with decreasing CD4+ cell count. However, in the HIV-PTB group, the ELISPOT responses had no discernible relationship to CD4+ cell count and were as likely to be elevated in subjects with severe immune depletion as they were to be elevated in subjects with higher CD4+ cell counts (figure 1B). When the ELISPOT frequency (spot-forming cells per 106 PBMCs) for each antigen was arbitrarily divided by the total CD4+ cell count, an interesting relationship became apparent (figure 1C). When considering all antigens, the median value of this ratio in the HIV-PTB group was 18.6 times higher than it was in the HIV-nonTB group (2.05 spot-forming cells per 106 PBMCs per mm3 [IQR, 0.61–6.39 spot-forming cells per 106 PBMCs per mm3] vs. 0.11 spot-forming cells per 106 PBMCs per mm3 [IQR, 0.018–0.38 spot-forming cells per 106 PBMCs per mm3]; P < .001, by Mann-Whitney U test).

Comparison of in vitro responses with the TST. Because there is no gold standard for the diagnosis of TB, estimation of sensitivity to detect TB has to rely on the response in overtly infected individuals (i.e., individuals with culture-positive or smear-positive disease). Using results from 21 individuals in the HIV-PTB group for whom the results of all 3 assays were available, we calculated the sensitivity of each method for detection of active TB (table 3). Of the patients in the HIV-PTB group, 57% had a TST response ⩾15 mm in diameter. Only a minor increase in sensitivity was gained with a cut-off value of 5 mm (with this cut-off value, 67% had a positive TST result, which was not a statistically significant increase; table 3). By contrast, both the ELISPOT (with a cut-off value of 25 spot-forming cells per 106 PBMCs, in agreement with extensive experience [28, 31, 32]) and the whole blood assay (with a cut-off value of 10 µg/mL) were more sensitive (P = .03 in both cases; table 3).

Table 3
View larger version:
Table 3

HIV-infected subjects with pulmonary tuberculosis exhibiting a positive response to tests for tuberculosis.

Therefore, on the basis that a positive response with the whole blood assay was indicative of TB infection, we estimated the prevalence of latent TB infection among HIV-infected individuals in our community and evaluated the decrease in sensitivity conferred by HIV infection on Mantoux TST and whole blood assay results for 69 subjects in the HIV-negative and HIV-nonTB groups for whom both whole blood assay and Mantoux TST responses were available (figure 2). Whole blood assay analysis indicated a prevalence of latent TB infection of 85% in the HIV-nonTB group (with a cut-off value of 10 µg/mL; the estimated prevalence was 79% with a cut-off value of 100 µg/mL) and a prevalence of 83% in the HIV-negative group, irrespective of the cut-off value. In the HIV-negative group, 92% of the subjects had a Mantoux TST response of >5 mm in diameter, and 69% had a response >15 mm in diameter (a percentage that accords well with the whole blood assay-based estimates of latent TB infection). By contrast, only 61% of HIV-nonTB subjects had a Mantoux TST response >5 mm in diameter (P = .004, compared with the HIV-negative group), and just 42% had a response >15 mm in diameter (P = .025). Taken together, these results suggest that the overall prevalence of latent TB infection in our study population was 80%–90% and that the Mantoux TST underestimates this prevalence in HIV-infected individuals irrespective of the cut-off value.

It follows from table 2 and figure 1C that a strongly positive whole blood assay result (or even a strongly positive Mantoux TST response) may help to distinguish active from latent TB in HIV-infected individuals, especially if the result is related to the CD4+ cell count. We explored this possibility by dividing results by CD4+ cell count (e.g., figure 1C) and then performing receiver-operator characteristic analysis on the transformed data, using the HIV-PTB group and the HIV-nonTB group as comparator groups. Significant results for area under the curve (AUC) analysis were obtained for the Mantoux TST (AUC, 0.6952; P = .011), and even more significant results were obtained for ESAT-6 (AUC, 0.832 ; P < .001) and CFP-10 (AUC, 0.891; P < .001) ELISPOT analysis. A ratio (defined by the sum of CFP-10 and ESAT-6 spot-forming cells per 106 PBMCs divided by the CD4+ cell count) of >1.0 predicted active TB in HIV-infected individuals with 88% sensitivity (95% CI, 69%–97%) and 80% specificity (95% CI, 52%–96%).

Previous SectionNext Section

Discussion

We have investigated M. tuberculosis-specific T cell IFN-γ release in HIV-infected and uninfected individuals without evidence of active TB and in HIV-infected patients with active pulmonary TB in a socioeconomically deprived urban township in Cape Town, South Africa. In HIV-infected individuals, the in vivo (TST) and in vitro (whole blood assay) response to PPD was markedly decreased (P < .001). By contrast, the response to M. tuberculosis antigens (ESAT-6, CFP-10, TB10.3, and Acr2) was not significantly decreased. Whole blood assay responses did not differ between the HIV-nonTB and HIV-PTB groups, but the IFN-γ ELISPOT response to ESAT-6, CFP-10, and PPD was higher in the HIV-PTB group (P ⩽ .04 in all cases), despite the fact that the HIV-PTB group had significantly lower CD4+ cell counts. A consequence of these contrasting trends was that a ratio of combined ESAT-6 and CFP-10 ELISPOT count divided by the CD4+ cell count of >1 had 88% sensitivity and 80% specificity for active pulmonary TB in HIV-infected individuals.

Although our whole blood assay performed as well as the ELISPOT for the overall detection of TB infection (table 3), was considerably easier to perform, and (as a consequence of dilution) required much less blood, it had poor specificity for the detection of active TB, even when results were stratified by CD4+ cell count. The result that underlies this difference between the overnight ELISPOT technique and the 3-day culture is the high ELISPOT response to CFP-10 and ESAT-6 in the HIV-PTB group (figure 1). We believe that this relates to bacillary burden, which was likely to be very high in our culture-positive or sputum-smear positive patients. A close relationship between the overnight IFN-γ response to these antigens, which are secreted by actively dividing bacilli, and the number of viable bacilli recovered from organs has been observed experimentally [33, 34]. Furthermore, the IFN-γ ELISPOT response to these antigens decreases during therapy for both active and latent TB in humans, an effect that presumably reflects a decreasing bacillary load [28, 35]. The ELISPOT assay primarily detects activated effector T cells [29], whereas the longer 3-day culture may permit the differentiation of—and, therefore, IFN-γ production by—memory T cells, the frequency of which is less influenced by antigen load.

Overall, our results support findings that ELISPOT analysis is less impaired by HIV infection than is the TST [7, 8]. Our analysis of these responses also suggests a simple way that the ELISPOT test could be incorporated into practice in the context of HIV infection. If the test has positive results and the sum of ESAT-6 and CFP-10 responses per 106 PBMCs divided by the CD4+ cell count is >1, this strongly suggests active disease that should be investigated and treated. If the ratio is <1, active disease is not excluded but, if disease is undetected, the patient should at least be treated for latent TB infection. We intended this study to be a pilot, and we acknowledge that the confidence limits of our analysis are wide and that the tests we used were “in house” assays. Our results support a larger prospective study of commercial tests that should be extended to the difficult diagnostic area of extrapulmonary TB, which is a common problem in HIV-infected individuals. It would be necessary to perform a simultaneous CD4+ cell count, but this requirement is not problematic in the developed world. In the developing world, the cost-effectiveness and feasibility of the test should also form a major component of evaluation.

Our results also indicate a very high prevalence of latent TB infection among people attending voluntary HIV counselling and testing facilities in Khayelitsha. The ∼80% prevalence of latent TB infection that we have detected in adults aged 30 years is consistent with estimates of the annual rate of infection made in nearby areas of Cape Town, where the incidence of TB is lower [36]. TB is very difficult to control in Khayelitsha, where the overall incidence of TB is 1612 cases per 100,000 population and where the rate of TB is much greater in the population of HIV-infected individuals [37]. The World Health Organization has called for extraordinary and unprecedented actions in such circumstances to reverse the rising tide of infection and disease. On the basis of this study, we are now considering mass isoniazid preventive therapy for all HIV-infected individuals without Mantoux TST. Because there is evidence from similar communities that individuals who transmit TB may, in fact, be oligosymptomatic chronic excretors [38, 39], consideration of targeted radiographic screening and greater use of isoniazid preventive therapy, irrespective of HIV-infection status, may also be pertinent. Our findings also underscore the need for novel preventive therapies, as well as a pre- or postinfection vaccine that might prevent reactivation and thereby prevent transmission [40].

Previous SectionNext Section

acknowledgments

Marcela Simsova of the Czech Academy of Sciences and Apronex (Prague, Czech Republic) was responsible for synthesis and purification of recombinant antigens. We are grateful to Virginia Azevedo and Dr. Helene Visser of the Health Department of the City of Cape Town (Cape Town, South Africa) and to Dr. Eric Goemaere of Medicins Sans Frontieres for constructive discussions prior to commencing the study. We thank Dr. Mark Nicol and Dr. Steven Lawn for their insightful review of the manuscript.

Financial support. Wellcome Trust of Great Britain (Ref. 072070) and the Medical Research Council of South Africa. This study was facilitated by a donation towards clinical facilities by the Rotary Club of Bad Oldesloe, Germany.

Potential conflicts of interest. All authors: no conflicts.

  • Received November 25, 2006.
  • Accepted February 28, 2007.
Previous Section
 

references

  1. ↵
    1. Corbett EL,
    2. Watt CJ,
    3. Walker N,
    4. et al
    . The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med 2003;163:1009-21.
    Abstract/FREE Full Text
  2. ↵
    1. Wood R,
    2. Maartens G,
    3. Lombard CJ
    . Risk factors for developing tuberculosis in HIV-1-infected adults from communities with a low or very high incidence of tuberculosis. J Acquir Immune Defic Syndr 2000;23:75-80.
    MedlineWeb of Science
  3. ↵
    1. Elliott AM,
    2. Hodsdon WS,
    3. Kyosiimire J,
    4. et al
    . Cytokine responses and progression to active tuberculosis in HIV-1-infected Ugandans: a prospective study. Trans R Soc Trop Med Hyg 2004;98:660-70.
    MedlineWeb of Science
  4. ↵
    1. Silveira H,
    2. Ordway D,
    3. Dockrell H,
    4. Jackson M,
    5. Ventura F
    . Cell-mediated immune responses to mycobacterial antigens in patients with pulmonary tuberculosis and HIV infection. Clin Exp Immunol 1997;110:26-34.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Elliott AM,
    2. Hurst TJ,
    3. Balyeku MN,
    4. et al
    . The immune response to Mycobacterium tuberculosis in HIV-infected and uninfected adults in Uganda: application of a whole blood cytokine assay in an epidemiological study. Int J Tuberc Lung Dis 1999;3:239-47.
    MedlineWeb of Science
  6. ↵
    1. Wallis RS,
    2. Vjecha M,
    3. Amir-Tahmasseb M,
    4. et al
    . Influence of tuberculosis on human immunodeficiency virus (HIV-1): enhanced cytokine expression and elevated beta 2-microglobulin in HIV-1-associated tuberculosis. J Infect Dis 1993;167:43-8.
    Abstract/FREE Full Text
  7. ↵
    1. Chapman AL,
    2. Munkanta M,
    3. Wilkinson KA,
    4. et al
    . Rapid detection of active and latent tuberculosis infection in HIV-positive individuals by enumeration of Mycobacterium tuberculosis-specific T cells. AIDS 2002;16:2285-93.
    CrossRefMedlineWeb of Science
  8. ↵
    1. Rangaka MX,
    2. Wilkinson KA,
    3. Seldon R,
    4. et al
    . Effect of HIV-1 infection on T-cell-based and skin test detection of tuberculosis infection. Am J Respir Crit Care Med 2007;175:514-20.
    Abstract/FREE Full Text
  9. ↵
    1. Mahairas GG,
    2. Sabo PJ,
    3. Hickey MJ,
    4. Singh DC,
    5. Stover CK
    . Molecular analysis of genetic differences between Mycobacterium bovis BCG and virulent M. bovis. J Bacteriol 1996;178:1274-82.
    Abstract/FREE Full Text
  10. ↵
    1. Sorensen AL,
    2. Nagai S,
    3. Houen G,
    4. Andersen P,
    5. Andersen AB
    . Purification and characterization of a low-molecular-mass T-cell antigen secreted by Mycobacterium tuberculosis. Infect Immun 1995;63:1710-7.
    Abstract/FREE Full Text
  11. ↵
    1. Berthet FX,
    2. Rasmussen PB,
    3. Rosenkrands I,
    4. Andersen P,
    5. Gicquel B
    . A Mycobacterium tuberculosis operon encoding ESAT-6 and a novel low-molecular-mass culture filtrate protein (CFP-10). Microbiology 1998;144:3195-203.
    Abstract/FREE Full Text
  12. ↵
    1. Lalvani A,
    2. Pathan AA,
    3. McShane H,
    4. et al
    . Rapid detection of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Am J Resp Crit Care Med 2001;163:824-8.
    Abstract/FREE Full Text
  13. ↵
    1. Liebeschuetz S,
    2. Bamber S,
    3. Ewer K,
    4. Deeks J,
    5. Pathan AA,
    6. Lalvani A
    . Diagnosis of tuberculosis in South African children with a T-cell-based assay: a prospective cohort study. Lancet 2004;364:2196-203.
    CrossRefMedlineWeb of Science
  14. ↵
    1. Nicol MP,
    2. Pienaar D,
    3. Wood K,
    4. 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 Infect Dis 2005;40:1301-8.
    Abstract/FREE Full Text
  15. ↵
    1. Lalvani A,
    2. Pathan AA,
    3. Durkan H,
    4. et al
    . Enhanced contact tracing and spatial tracking of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Lancet 2001;357:2017-21.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Ewer K,
    2. Deeks J,
    3. Alvarez L,
    4. et al
    . Comparison of T-cell-based assay with tuberculin skin test for diagnosis of Mycobacterium tuberculosis infection in a school tuberculosis outbreak. Lancet 2003;361:1168-73.
    CrossRefMedlineWeb of Science
  17. ↵
    1. Mazurek GH,
    2. Jereb J,
    3. Lobue P,
    4. Iademarco MF,
    5. Metchock B,
    6. Vernon A
    . Guidelines for using the QuantiFERON-TB Gold test for detecting Mycobacterium tuberculosis infection, United States. MMWR Recomm Rep 2005;54:49-55.
    Medline
  18. ↵
    1. Johnson JL,
    2. Nyole S,
    3. Okwera A,
    4. et al
    . Instability of tuberculin and Candida skin test reactivity in HIV-infected Ugandans. The Uganda-Case Western Reserve University Research Collaboration. Am J Respir Crit Care Med 1998;158:1790-6.
    Abstract/FREE Full Text
  19. ↵
    1. StopTBpartnership
    . The global plan to stop TB: Executive summary. 2006. Available at: http://www.stoptb.org/globalplan/plan_main.asp. Accessed 6 May 2007.
  20. ↵
    1. Scholvinck E,
    2. Wilkinson KA,
    3. Whelan AO,
    4. Martineau AR,
    5. Levin M,
    6. Wilkinson RJ
    . Gamma interferon-based immunodiagnosis of tuberculosis: comparison between whole-blood and enzyme-linked immunospot methods. J Clin Microbiol 2004;42:829-31.
    Abstract/FREE Full Text
  21. ↵
    1. Wilkinson KA,
    2. Simsova M,
    3. Scholvinck E,
    4. et al
    . Efficient ex vivo stimulation of Mycobacterium tuberculosis-specific T cells by genetically detoxified Bordetella pertussis adenylate cyclase antigen toxoids. Infect Immun 2005;73:2991-8.
    Abstract/FREE Full Text
  22. ↵
    1. Skjot RL,
    2. Brock I,
    3. Arend SM,
    4. et al
    . Epitope mapping of the immunodominant antigen TB10.4 and the two homologous proteins TB10.3 and TB12.9, which constitute a subfamily of the esat-6 gene family. Infect Immun 2002;70:5446-53.
    Abstract/FREE Full Text
  23. ↵
    1. Hogarth PJ,
    2. Logan KE,
    3. Vordermeier HM,
    4. Singh M,
    5. Hewinson RG,
    6. Chambers MA
    . Protective immunity against Mycobacterium bovis induced by vaccination with Rv3109c—a member of the esat-6 gene family. Vaccine 2005;23:2557-64.
    CrossRefMedlineWeb of Science
  24. ↵
    1. Wilkinson KA,
    2. Stewart GR,
    3. Newton SM,
    4. et al
    . Infection biology of a novel α-crystallin of Mycobacterium tuberculosis: Acr2. J Immunol 2005;174:4237-43.
    Abstract/FREE Full Text
  25. ↵
    1. Stewart GR,
    2. Newton SM,
    3. Wilkinson KA,
    4. et al
    . The stress-responsive chaperone alpha-crystallin 2 is required for pathogenesis of Mycobacterium tuberculosis. Mol Microbiol 2005;55:1127-37.
    CrossRefMedlineWeb of Science
  26. ↵
    1. Mohammed A,
    2. Ehrlich R,
    3. Wood R,
    4. Cilliers F,
    5. Maartens G
    . Screening for tuberculosis in adults with advanced HIV infection prior to preventive therapy. Int J Tuberc Lung Dis 2004;8:792-5.
    MedlineWeb of Science
  27. ↵
    1. Mosimaneotsile B,
    2. Talbot EA,
    3. Moeti TL,
    4. et al
    . Value of chest radiography in a tuberculosis prevention programme for HIV-infected people, Botswana. Lancet 2003;362:1551-2.
    CrossRefMedlineWeb of Science
  28. ↵
    1. Wilkinson KA,
    2. Kon OM,
    3. Newton SM,
    4. et al
    . Effect of treatment of latent tuberculosis infection on the T cell response to Mycobacterium tuberculosis antigens. J Infect Dis 2006;193:354-9.
    Abstract/FREE Full Text
  29. ↵
    1. Wilkinson KA,
    2. Wilkinson RJ,
    3. Pathan A,
    4. et al
    . Ex vivo characterization of early secretory antigenic target 6-specific T cells at sites of active disease in pleural tuberculosis. Clin Infect Dis 2005;40:184-7.
    Abstract/FREE Full Text
  30. ↵
    GraphPad software Web site. Available at: http://www.graphpad.com. Accessed 25 April 2007.
  31. ↵
    1. Lalvani A
    . Counting antigen-specific T cells: a new approach for monitoring response to tuberculosis treatment? Clin Infect Dis 2004;38:757-9.
    FREE Full Text
  32. ↵
    1. Jeffries DJ,
    2. Hill PC,
    3. Fox A,
    4. et al
    . Identifying ELISPOT and skin test cut-offs for diagnosis of Mycobacterium tuberculosis infection in The Gambia. Int J Tuberc Lung Dis 2006;10:192-8.
    MedlineWeb of Science
  33. ↵
    1. Vordermeier HM,
    2. Chambers MA,
    3. Cockle PJ,
    4. Whelan AO,
    5. Simmons J,
    6. Hewinson RG
    . Correlation of ESAT-6-specific gamma interferon production with pathology in cattle following Mycobacterium bovis BCG vaccination against experimental bovine tuberculosis. Infect Immun 2002;70:3026-32.
    Abstract/FREE Full Text
  34. ↵
    1. Dietrich J,
    2. Aagaard C,
    3. Leah R,
    4. et al
    . Exchanging ESAT6 with TB10.4 in an Ag85B fusion molecule-based tuberculosis subunit vaccine: efficient protection and ESAT6-based sensitive monitoring of vaccine efficacy. J Immunol 2005;174:6332-9.
    Abstract/FREE Full Text
  35. ↵
    1. Lalvani A,
    2. Nagvenkar P,
    3. Udwadia Z,
    4. et al
    . Enumeration of T cells specific for RD1-encoded antigens suggests a high prevalence of latent Mycobacterium tuberculosis infection in healthy urban Indians. J Infect Dis 2001;183:469-477.
    Abstract/FREE Full Text
  36. ↵
    1. Verver S,
    2. Warren RM,
    3. Munch Z,
    4. et al
    . Transmission of tuberculosis in a high incidence urban community in South Africa. Int J Epidemiol 2004;33:351-7.
    Abstract/FREE Full Text
  37. ↵
    1. Coetzee D,
    2. Hildebrand K,
    3. Boulle A,
    4. et al
    . Outcomes after two years of providing antiretroviral treatment in Khayelitsha, South Africa. AIDS 2004;18:887-95.
    CrossRefMedlineWeb of Science
  38. ↵
    1. Corbett EL,
    2. Charalambous S,
    3. Moloi VM,
    4. et al
    . Human immunodeficiency virus and the prevalence of undiagnosed tuberculosis in African gold miners. Am J Respir Crit Care Med 2004;170:673-9.
    Abstract/FREE Full Text
  39. ↵
    1. Lawn SD,
    2. Bekker LG,
    3. Middelkoop K,
    4. Myer L,
    5. Wood R
    . Impact of HIV infection on the epidemiology of tuberculosis in a peri-urban community in South Africa: the need for age-specific interventions. Clin Infect Dis 2006;42:1040-7.
    Abstract/FREE Full Text
  40. ↵
    1. Young D,
    2. Dye C
    . The development and impact of tuberculosis vaccines. Cell 2006;124:683-7.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. Clin Infect Dis. (2007) 44 (12): 1639-1646. doi: 10.1086/518234
  1. AbstractFree
  2. » Full Text (HTML)
  3. Full Text (PDF)

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most