domingo, 24 de abril de 2011

Biomarcador para tuberculose

Review

Nature Reviews Immunology 11, 343-354 (May 2011) | doi:10.1038/nri2960

Immunological biomarkers of tuberculosis

Gerhard Walzl1, Katharina Ronacher1, Willem Hanekom2, Thomas J. Scriba2 & Alimuddin Zumla3  About the authors
top
Currently there are no sufficiently validated biomarkers to aid the evaluation of new tuberculosis vaccine candidates, the improvement of tuberculosis diagnostics or the development of more effective and shorter treatment regimens. To date, the detection of Mycobacterium tuberculosis or its products has not been able to adequately address these needs. Understanding the interplay between the host immune system and M. tuberculosis may provide a platform for the identification of suitable biomarkers, through both unbiased and targeted hypothesis-driven approaches. Here, we review immunological markers, their relation to M. tuberculosis infection stages and their potential use in the fight against tuberculosis.
Tuberculosis is caused by infection with Mycobacterium tuberculosis, which is transmitted through inhalation of aerosolized droplets. Tuberculosis constitutes a serious global health problem with nearly 10 million new cases of tuberculosis and 1.7 million deaths every year1. Furthermore, it is estimated that two billion people live with latent M. tuberculosisinfection and represent a potential source of future active tuberculosis2. Global control of tuberculosis can only be achieved through the development of effective vaccines, improved diagnostics, and novel and shortened therapy regimens3.
The need for tuberculosis biomarkers arises, in part, from the lack of suitable tests to detect M. tuberculosis or its products in host samples. The most widely used diagnostic test is microscopic detection of acid-fast bacilli in sputum (referred to here as the sputum smear test), which has a sensitivity of 34–80%4. Although the sputum culture test is more sensitive than the sputum smear test, it can take weeks to obtain results from the sputum culture test and laboratory facilities for this test are often not available in high incidence areas. A recently developed ex vivo M. tuberculosis gene amplification test (GeneXpert MTB/RIF) can be used to diagnose tuberculosis and can also detect resistance to rifampicin, one of the main antibiotics used in tuberculosis treatment, which serves as a marker for multidrug resistance. This test allows automated sample processing and produces results within two hours with excellent sensitivity5. However, its use is restricted to the identification of active pulmonary tuberculosis, as it cannot detect latent disease.
Host biomarkers are therefore needed to help to diagnose tuberculosis, to providecorrelates of risk of tuberculosis and correlates of protection against active disease, and to determine the response to therapy. This Review discusses potential host immunological biomarkers for M. tuberculosis exposure, infection, disease and treatment and correlates these with stages of host–pathogen interaction.

Pathogenesis of tuberculosis

The interactions of M. tuberculosis with its host are complex, and our understanding of pathogenesis and of the protective immune responses during infection is constantly changing as technology advances. The recent observation of marked heterogeneity of the host immune responses6, and possibly even of bacterial metabolism, within the same individual7 has important implications for biomarker identification. The early conclusion from these studies is that latent and active tuberculosis do not represent two separate and distinct states, but that a continuum of host–pathogen interactions exists7.
Figure 1 summarizes some current concepts of the clinical phenotypes, pathogenesis and host immune responses in tuberculosis and the resulting opportunities for biomarker discovery. Not all individuals who are exposed to M. tuberculosis become infected, suggesting that some may clear the bacteria through innate immune mechanisms; however, this has not been proven. In those individuals who become infected, the pathogen may be cleared through adaptive immune mechanisms, involving different T cell populations. These populations include effector memory T cells (which may be transiently present in the blood if bacteria are cleared) and central memory T cells (which may remain for life but may not provide protection in all individuals). Failure of the immune system to eradicate the bacteria may result in a spectrum of host–pathogen interactions. The bacteria may exist in a quiescent state for prolonged periods (referred to as latent tuberculosis); however, the bacteria may ultimately start replicating and escape immune control, resulting in clinically active tuberculosis.
Figure 1 | Immune responses and potential host biomarkers of Mycobacterium tuberculosis exposure and infection.
Figure 1 : Immune responses and potential host biomarkers of Mycobacterium tuberculosis exposure and infection. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.comExposure to Mycobacterium tuberculosis (Mtb) can result in different clinical outcomes, which include the absence of any clinical or laboratory evidence of infection, infection without clinically active disease and active disease. The stage of infection is determined by the ability of the host innate and adaptive immune systems to eradicate or control M. tuberculosis. A wide range of specific and nonspecific host immune responses contribute to the differential outcomes of exposure and infection, although there is a lack of detailed understanding of the underlying mechanisms. The host response at the different infection stages represents opportunities to measure individual markers or combinations of markers (biosignatures) that have diagnostic or prognostic potential, although they may have to be interpreted in a specific clinical context. The first three phases ofM. tuberculosis infection are asymptomatic. Only some patients with tuberculosis develop immunological signs of infection; others can eliminate the bacteria during the innate immune phase, without generating T cell memory (and thus have negative tuberculin skin test (TST) and interferon-γ release assay (IGRA) results). In the adaptive immune phase, T cells are engaged by antigen-presenting cells, and this generates effector and memory T cells (both effector memory T (TEM) and central memory T (TCM) cells). B cells are also activated and M. tuberculosis-specific antibodies are produced. The infection may be cleared at this stage. However, most exposed individuals will enter the quiescent phase, which may persist for life. In this phase, the bacteria are contained inside granulomas, which consist of a central area, containing infected macrophages and giant foam cells, surrounded by TCM and B cells. Although the host fails to eradicate the pathogen, replication and dissemination of the bacteria are prevented. An optimal T helper (TH) cell balance is required to control M. tuberculosis while limiting immunopathology. This balanced reaction includes pro-inflammatory TH1-type responses (characterized by interferon-γ (IFNγ), tumour necrosis factor (TNF) and interleukin-12 (IL-12) production) and TH17-type responses (characterized by IL-17 production). However, it also involves TH2-type responses (associated with IL-4 production) and regulatory T (TReg) cell phenotypes that limit immunopathology. The replicating phase is symptomatic and at this stage the bacteria have escaped immune control. Granulomas are disrupted, the acute-phase response is activated and the levels of pro-inflammatory markers increase. Enhanced immunosuppression becomes evident as TH cell balance is disturbed, and memory T cell populations and antibody production may change. HIV co-infection and associated CD4+ T cell depletion can trigger this immune escape by M. tuberculosis. CXCL10, CXC-chemokine ligand 10; TCR, T cell receptor.

Only up to 10% of humans infected with M. tuberculosis will progress to active tuberculosis disease during their lifetime8. M. tuberculosis has evolved elaborate survival mechanisms in humans, allowing the bacterium to remain in a clinically inactive state, although it constantly engages with the human immune system. In this state, the immune response prevents active replication but fails to eradicate the bacteria. Any subsequent weakening of the host immune system may result in clinical disease. The mechanisms for these differential outcomes to M. tuberculosis exposure remain unclear.
Live M. tuberculosis may also persist after an initially successful treatment of active tuberculosis, and dynamic interchanges between host and pathogen determine whetherrelapse to active disease will occur (Fig. 2). These interchanges may be particularly relevant in immunocompromised individuals; in sub-Saharan Africa a high proportion of HIV-infected patients are co-infected with M. tuberculosis, and the HIV-induced immunodeficiency contributes to the increasing numbers of tuberculosis cases and deaths9.
Figure 2 | Differential outcomes of tuberculosis treatment are associated with different infection phases.
Figure 2 : Differential outcomes of tuberculosis treatment are associated with different infection phases. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.comTreatment of active tuberculosis results in clinical cure in the majority of patients, although drug-resistant organisms and poor treatment adherence can result in treatment failure. Clinical cure is characterized by negative bacteriological examination forMycobacterium tuberculosis and by resolution or improvement of symptoms and changes in chest X-ray examinations. The term is used to refer to both sterilizing cure and a return to the quiescent phase (non-sterilizing cure). However, sterilizing cure cannot be reliably differentiated from non-sterilizing cure, although the persistence of effector memory T (TEM) cells may point towards the continued presence of live, non-replicating bacteria. TEM cells home to peripheral tissue, can differentiate directly into effector cells and are detectable by measuring interferon-γ (IFNγ) production in short-term assays using whole blood or peripheral blood mononuclear cells stimulated with M. tuberculosis antigens. Central memory T (TCM) cells are long-lived, even in the absence of persistent antigen, and home to secondary lymphoid organs, require longer term stimulation assays and signify previous immunological sensitization to the pathogen. Active tuberculosis can recur, either through reinfection with a new bacterial strain (in patients whose tuberculosis treatment resulted in sterilizing cure) or through relapse with the original bacterial strain (in patients whose infection returned to a quiescent phase after treatment). Measurements of memory T cell subpopulations and other biomarkers for pathogen persistence have so far not been adequately investigated for their ability to predict treatment outcome, and the field relies on clinical evidence of mycobacterial activity.

So, the traditional views of M. tuberculosis infection as either a latent or active disease have been replaced by a model where a continuum of host–pathogen interactions exists, resulting in a spectrum of immune responses7.

Immunology of tuberculosis

Protective immunity against M. tuberculosis is not completely understood but depends on a wide range of innate and adaptive immune mechanisms. T cell-mediated immune responses are important in the host control of M. tuberculosis infection. The ability of CD4+ T cells to produce interferon-γ (IFNγ), which activates phagocytes to contain the intracellular pathogen, is central in protection. Indeed, T helper 1 (TH1) cells and the IFNγ that they produce are crucial for protection against disease. This is evident from the increased risk of tuberculosis in individuals with deficiencies in their IFNγ and interleukin-12 (IL-12; which promotes TH1 cell differentiation) signalling pathways10, and from the association between CD4+ T cell depletion and elevated susceptibility to tuberculosis in HIV-infected individuals11.
Many other CD4+ T cell subsets, in addition to IFNγ-producing TH1 cells, may also have a role; for example, IL-17-producing CD4+ T cells were shown to mediate the recruitment of protective TH1 cells to the lung upon M. tuberculosis challenge12. Induction of IL-17 expression following vaccination of cattle using a prime–boost vaccine strategy also correlated with protection against Mycobacterium bovis tuberculosis13. Furthermore, increased frequencies of regulatory CD4+ T (TReg) cells during active disease may ensure that the TH1 cell response is not excessive, and this would help minimise lung damage in tuberculosis14. This control of overactive inflammatory responses may also be achieved by inhibitory molecules, such as programmed cell death 1 (PD1), that are expressed by the effector T cells themselves, and mice that lack expression of PD1 develop more severe disease following M. tuberculosis challenge15. However, too much TReg cell activity may suppress protective inflammatory responses. Therefore, the critical determinant of protection may be a balanced and well-regulated immune response.
The CD8+ T cell response to M. tuberculosis is normally of a lower magnitude than the CD4+ T cell response; however, CD8+ T cells may modulate phagocyte activity or produce molecules such as granulysin that may be directly cytotoxic to the mycobacteria16, 17. Similarly, other cytokines, in addition to IFNγ, may also be crucial; for example, tumour necrosis factor (TNF) is important for establishing the granuloma16, which is a well-organized collection of innate and adaptive cells that forms to contain the pathogen.
Furthermore, the importance of innate immunity cannot be overemphasized. The dendritic cells or macrophages that engulf M. tuberculosis in the lung can interact with the pathogen through numerous receptors, which can induce distinct innate immune responses18. These responses may even eradicate the bacterium before the establishment of adaptive immunity. However, bacterial evasion of innate cellular responses, leading to delayed induction of the adaptive immune system19, appears to be a more characteristic outcome of M. tuberculosis infection and allows the pathogen to become established at the site of disease. Natural killer (NK) cells and granulocytes are also thought to have a role in protection.
Our knowledge of protective immunity against tuberculosis remains incomplete; a classical example is that the role of B cells remains undefined, even though these cells are found in substantial numbers in granulomas20. A list of immunological markers that differentiate between different outcomes of M. tuberculosis infection is shown in Table 1.

Correlates of risk and protection

Knowledge about the correlates of tuberculosis risk and protection will facilitate rapid screening of new tuberculosis vaccine candidates and targeted intervention to prevent tuberculosis disease. A non-placebo-controlled clinical trial (for example, following M. bovis bacillus Calmette–Guérin (BCG) vaccination or M. tuberculosis infection) allows for the delineation of correlates of risk of tuberculosis disease, by comparing the immune responses of those who ultimately develop tuberculosis with those who do not. By contrast, correlates of protection can be identified only in a randomized controlled clinical trial of an effective vaccine21, in which immune responses are compared between protected individuals in the vaccine and placebo groups. This is due to the fact that there is no evidence that latent infection induces protection against the development of active tuberculosis, rather it places such people at risk of tuberculosis. Therefore, the immune responses in latently infected individuals who never develop tuberculosis (who account for >90% of latently infected individuals) are not useful as correlates of protection. It has been proposed that correlates of risk of tuberculosis disease could guide future exploration of correlates of protection. Unfortunately, no validated clinical correlate of either risk or protection currently exists.
The tuberculin skin test reaction in individuals who have received the BCG vaccine correlates poorly with subsequent development of active disease22. Therefore, in clinical trials of new tuberculosis vaccines, the vaccine response (immunogenicity) is now commonly determined by measuring specific CD4+ T cell responses, particularly the production of TH1 cell-associated cytokines such as IFNγ. However, Kagina et al. recently reported that specific CD4+ T cell responses 10 weeks after BCG vaccination of newborns do not correlate with ultimate risk of tuberculosis disease23. Importantly, CD4+ T cell expression of IFNγ, co-expression of IFNγ, TNF and IL-2 (by polyfunctional CD4+ T cells) or CD8+ T cell responses did not correlate with risk of disease during the first 2 years of life. By contrast, antigen-specific polyfunctional T cells in animal models have been correlated with protection against intracellular pathogens such as Leishmania major24, and studies in macaques have suggested an importance for CD8+ T cells in the control ofM. tuberculosis25.
These results suggest that the immune responses that are crucial for protection against clinically active M. tuberculosis infection may not necessarily translate into correlates of protection or risk in humans. Furthermore, the failure to identify such correlates may highlight the limitations of sampling peripheral blood instead of the site of infection (the lung); however, the procedures used to obtain lung samples can only be performed in highly specialized facilities. CD4+ T cell responses at this mucosal site, including those measured in local lymph nodes, have shown the best correlation with protection in mice26. There are also multiple examples of antigen-specific CD4+ T cell responses in the spleen of mice that have correlated with protection27, and these studies also highlighted the importance of measuring responses at an optimal post-vaccination time point. It is noteworthy that an association between CD4+ T cell response and protection against tuberculosis in mice has not been a universal finding; other studies suggest that IFNγ levels may simply be a measure of inflammatory status28. This is also supported by human studies of tuberculosis disease, and it emphasizes that protection cannot be defined by peripheral blood TH1 cell responses alone.
Assays to determine the inhibition of M. tuberculosis growth evaluate both T cell and innate immune cell functions. Bacterial growth is inhibited when M. tuberculosis is incubated with peripheral blood mononuclear cells (PBMCs) or whole blood from BCG-vaccinated adults, but not during incubation with cells from non-vaccinated individuals29. As mentioned above, multiple animal studies have suggested important roles for various subsets of T cells, such as TH17 cells, in addition to TH1 cells. Moreover, a well-orchestrated, balanced immune response, rather than simply an effector or memory TH1 cell response, may be vital for protection. For example, cynomolgus monkeys that developed tuberculosis disease following low-dose aerosol infection had lower numbers of CD4+ TReg cells in the bronchoalveolar lavage fluid than monkeys that did not progress to disease30. Furthermore, γδ T cells31 and T cells restricted by non-classical molecules — such as CD1, HLA-E and MR1 (MHC class I-related protein) — also have a role in protection and should be included in hypothesis-driven approaches for describing immune correlates of protection.
Human immune responses to M. tuberculosis and tuberculosis vaccines, on both individual and population levels, are characterized by marked heterogeneity. For example, Malawian BCG recipients produce significantly less M. tuberculosis-specific IFNγ than British vaccinees6, suggesting a role for genetic and/or environmental factors (such as environmental mycobacterial exposure) in the immune response to BCG. This raises the possibility that identified biomarkers may not apply to all individuals within a population or across different populations. Given this hurdle, and the limitations of working primarily with peripheral blood rather than with material from the site of disease, we propose that new approaches to identify correlates of risk and protection should account for host heterogeneity and not be limited to hypotheses that focus primarily on T cells. Ongoing analyses of large clinical cohorts that aim to find correlates of risk for tuberculosis disease have therefore focused on data-driven methods, such as global screens with 'omics' approaches, including transcriptomics (see below).
Taken together, these studies suggest that the immune responses that have so far been considered to be crucial for protection against tuberculosis, including IFNγ production and CD4+ T cell responses, are not sufficient for protection and do not represent usable correlates of risk or protection in the context of vaccine trials.

Host markers for active tuberculosis

Early accurate diagnosis of active tuberculosis disease is important to stop transmission, but current diagnostic tests are inadequate32. We continue to rely on the century-old sputum smear test in most high-tuberculosis-endemic countries. Furthermore, specific challenges are associated with the diagnosis of extrapulmonary tuberculosis, sputum smear-negative tuberculosis (active pulmonary tuberculosis with less than 10,000 bacilli per ml of sputum) and childhood tuberculosis.
Antigen-induced IFNγ screening for active and latent tuberculosis. For the past century, the tuberculin skin test has been the only available screening test for latent tuberculosis infection, but this method cannot differentiate between M. tuberculosis and environmental mycobacterial infection. A decade ago, T cell-based IFNγ release assays (IGRAs) were developed, which assess IFNγ production after in vitro stimulation of whole blood or PBMCs with M. tuberculosis-specific immunodominant antigens, such as 6 kDa early secretory antigenic target (ESAT6), 10 kDa culture filtrate antigen (CFP10) and TB7.7 (Refs 33, 34). These assays have now become the gold standard for the identification of sensitization to M. tuberculosis35. However, evaluation of IGRA results for the detection of latent tuberculosis has been difficult owing to the absence of a gold standard for latency.
A meta-analysis of these studies showed that IGRAs are as sensitive as and more specific than the tuberculin skin test36. However, a multi-centre Phase III clinical study demonstrated that IGRAs are unsuitable for diagnosing active disease, particularly in high-tuberculosis-endemic areas37. Nevertheless, some studies suggest that the response detected by IGRAs, when quantified, is stronger in patients with active tuberculosis than in those with latent tuberculosis38. Furthermore, IGRAs performed on T cells isolated from the site of tuberculosis disease (for example, pleural effusions or cerebrospinal fluid) were found to be highly sensitive and specific39, 40. In addition, multi-cytokine biosignatures may differentiate between active or latent tuberculosis41; the expression levels of mRNA transcripts encoding CXC-chemokine ligand 8 (CXCL8; also known as IL-8), the TReg cell-associated transcription factor forkhead box P3 (FOXP3) and IL-12β following ESAT6 stimulation of PBMCs hold promise in this regard42. Similarly, Djoba et al. have shown that cytokine expression levels in the blood can differentiate between pulmonary and pleural tuberculosis43.
It has also been suggested that distinct cytokine expression profiles of CD4+ T cells are associated with the bacterial loads of different infection states44. Latent infection is dominated by the presence of T cells secreting IL-2 only or IFNγ and IL-2, whereas T cells secreting only IFNγ are more frequent during active disease45. More recently, M. tuberculosis-specific T cells secreting only TNF have been found to be more frequent in individuals with active tuberculosis, and this may have an application as a new diagnostic test for active disease versus latent infection46.
Serological biomarkers for active tuberculosis. Serological tests based on the detection of circulating antibodies against M. tuberculosis-specific antigens have several advantages, as they are simple, cheap and feasible for point-of-care diagnostics. However, a comparative study of 19 commercially available tests found sensitivities ranging from 0.09% to 59.7%47. Specificities ranged from 53% to 98.7%, and the tests with high specificities frequently exhibited poor sensitivity. Potential M. tuberculosisantigen targets were reviewed in a recent meta-analysis by Steingart et al.48. A total of 254 studies were identified, encompassing 9 native and 27 recombinant proteins, 15 lipid-derived antigens and an additional 30 combined antigen targets, but low sensitivity and specificity limits the use of these serological tests.
A recent study used a systems immunology approach involving high-throughput screening to detect antibodies to the entire M. tuberculosis proteome in sera from 500 patients. The results showed that antibody responses correlated with bacterial burden and that a small pool of antigens is recognized during active tuberculosis, suggesting a diagnostic potential for these antibodies49. Therefore, combinations of host markers, including serological markers, need to be evaluated further to identify biosignatures with diagnostic potential for active disease.

Biomarkers for treatment response

The different outcomes of antituberculous chemotherapy and the underlying continuum of infection phases are summarized in Fig. 2. Biomarkers for the outcome of tuberculosis treatment are most urgently needed both for clinical decision making and to facilitate the shortening of clinical trials of new antituberculosis drugs or regimens. Current tuberculosis treatment strategies and the role of treatment response markers in developing new regimens are summarized in Fig. 3.
Figure 3 | Tuberculosis treatment and the potential role of biomarkers in clinical decision making and clinical trials.
Figure 3 : Tuberculosis treatment and the potential role of biomarkers in clinical decision making and clinical trials. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.comPatients diagnosed with a first episode of tuberculosis are treated for 2 months with four drugs — isoniazid, rifampicin, pyrazinamide and ethambutol; collectively known as HRZE — and subsequently with isoniazid and rifampicin (HR) for 4 months69. Patients with recurrent tuberculosis have an increased risk for drug-resistant tuberculosis and are treated for even longer periods, and drug-resistant tuberculosis is treated for 20 months or longer (not shown). The aim is to eventually develop 2-month treatment regimens, but the current drug combinations have high relapse rates if treatment duration is shorter than 6 months. Baseline testing for drug susceptibility is recommended and this is also performed after 2 months of treatment if sputum smear or culture tests remain positive. However, the results of these tests at month 2 have poor individual predictive ability for ultimate treatment outcome. Differentiating markers are currently being investigated at baseline, early during treatment (in the first 2 months), later during treatment (up to month 6) and after treatment completion. Qualified biomarkers are needed to accelerate the development of new, shorter treatment regimens, to guide clinical decision making for individual patients and to increase our understanding of the biology of host–pathogen interactions. Baseline markers may allow the stratification of patients into different treatment arms, based on the extent of their disease and the risk for poor treatment response. Early treatment response markers will give an earlier indication of response to chemotherapy than sputum tests and will alert clinicians to the potential presence of drug resistance, poor treatment adherence or ineffective drugs in clinical trials. Late and end of treatment markers are needed to indicate sterilizing cure or persistence of live, albeit non-replicating, bacteria, which may subsequently lead to relapse. The development and subsequent implementation of future shorter treatment regimens will be greatly aided by qualified biomarkers at baseline, during early treatment and at the end of treatment.

Pretreatment markers with predictive ability. Baseline biomarkers could identify the requirements of individual patients for specific treatment regimens. Individuals with high bacterial burden and extensive inflammation may require longer treatment regimens than individuals with minimal disease. For clinical trials, such biomarkers could ensure standardization across treatment groups and thereby reduce the required study size. High baseline bacterial load, measured by time-to-detection in liquid culture, low body mass index and more extensive disease on chest X-rays at diagnosis have been shown to predict relapse following treatment50, 51. However, results for baseline bacterial load tests take several days or even weeks, and X-rays are not universally available and their assessment is difficult to standardize.
The decision to shorten treatment time cannot just be based on the presence or absence of cavities on chest X-rays at the time of diagnosis and on early treatment response. Indeed, Johnson et al. found an increased relapse rate when tuberculosis treatment was reduced from 6 to 4 months in patients without chest X-ray cavities and whose sputum culture test was negative 2 months after treatment initiation51. Although none of the participants in this study had cavities, the individuals who suffered a relapse had more extensive lung damage and a higher bacterial burden on initial diagnosis than the patients who were cured. This suggests that more sophisticated measures of disease severity at diagnosis and of treatment response may be required to guide treatment regimens. Therefore, host immunological markers that correlate with extent of disease and may indicate a risk for relapse at baseline are currently being investigated.
Host markers during early treatment. Currently, the earliest measures of the effect of treatment are sputum smear or sputum culture conversion from positive at baseline to negative 2 months after starting therapy. This time interval of 2 months is a prohibitive delay for clinical management and for clinical trials of new drugs. Ineffective treatment allows unchecked bacterial replication and ongoing tissue destruction, spread of bacteria to other tissues and even the development of drug resistance in bacteria exposed to suboptimal drug combinations. Combinations of host immunological and clinical markers that are better indicators of early response to treatment and that reflect restoration of the balance of pro- and anti-inflammatory responses may therefore be useful, possibly in conjunction with markers of the extent of disease (Table 1).
Although host pro-inflammatory signatures may not be specific for tuberculosis, they may still be very useful if interpretation is performed within a specific clinical context or in conjunction with tuberculosis-specific markers.
IGRAs have not proven useful for monitoring treatment success52, although the levels of IFNγ in PBMC stimulation assays using recombinant M. bovis BCG 32 kDa protein (also known as Ag85A) increase during treatment of tuberculosis. By contrast, IL-10 levels decrease, and the IFNγ/IL-10 ratio correlates with treatment success and can also distinguish between active and latent tuberculosis53. Similarly, the ratio of IL-4 and its antagonistic splice variant IL-4δ2 increases during tuberculosis treatment, and changes in the IL-4/IL-4δ2 ratio occur early during tuberculosis treatment and may predict subsequent outcome54.
End of treatment markers for relapse. Shorter treatment regimens are a major goal of new drug development, but relapse will be the main risk of such new therapies. As relapse usually occurs within 2 years of treatment completion, any study evaluating shortened regimens would have to follow up cured tuberculosis patients for more than two years, and this would substantially increase the complexity and cost of the study. Therefore, a marker of sterilizing cure would answer an important clinical and drug trial need.
After tuberculosis treatment, the frequencies of certain T cell populations — T cells secreting IFNγ only, those secreting both IFNγ and IL-2 (Ref. 55) and IFNγ-, IL-2- and TNF-secreting T cells44 — change to the frequencies found in latently infected individuals. The presence of M. tuberculosis-specific effector memory T cell responses after a previous episode of spontaneously cured active tuberculosis (that occurred in the pre-antibiotic era, for example) may indicate the persistence of live bacteria and a return to a stable, quiescent phase of infection. By contrast, some individuals retain only central memory T cells, suggesting that the bacteria have been eradicated56. No data are available for the persistence of memory T cells as predictors of relapse. However, as a quiescent phase may subsequently revert to a phase of bacterial replication and relapse with active tuberculosis the implications of memory T cell phenotypes at the end of treatment should be further investigated.
In sum, certain immune markers hold promise as potential treatment biomarkers. These could replace the current baseline measures of the extent of disease, as well as the earliest treatment response markers at month 2, and could also help to identify markers for relapse.

A new direction for biomarker discovery

Owing to our limited knowledge of host immune responses to M. tuberculosis infection and the resultant lack of adequate biomarkers for the different phases of infection, the use of 'omics' approaches may be needed for biomarker discovery. Notably, several recent studies of this type have reported encouraging results57, 58, 59, 60.
Transcriptomic signatures. Transcriptomic studies from high endemic areas of tuberculosis have identified signatures involving host gene expression profiles that differentiate between different M. tuberculosis infection states. The first study used unique transcriptional features to identify nine RNA transcripts from whole blood, and these were used to successfully differentiate between patients with active, cured and recurrent tuberculosis, as well as latently infected individuals60 (Table 1). Similarly, Maertzdorf et al.61 identified unique transcriptional profiles that could distinguish between active tuberculosis, latent infection and uninfected donors. High-affinity IgG Fc receptor IB (FcγRIB) was the most differentially expressed gene and, together with four other transcripts, could discriminate between active disease and latent infection. Genes related to the innate immune response were overexpressed in active tuberculosis, whereas genes related to apoptosis and NK cell activity were upregulated during latent tuberculosis. Another study reported that a minimal group of genes (lactotransferrin,FCGR1A (also known as CD64) and RAB33A) was sufficient for classification of uninfected and latently infected individuals and patients with active tuberculosis. RAS and RAB interactor 3 (RIN3) was also identified as important when comparing active, recurrent, cured and latent tuberculosis60.
Molecular profiling in heterogeneous tissues, such as blood, is confounded by the relative proportions of different cell types in such tissues. Separation of tissues into pure cell populations by antibody-based methods or by microdissection would be appropriate but is not always feasible. An in silico deconfounding approach may offer an alternative to cell-specific molecular profiling. In this approach, normalization of gene expression data takes into account the number of cell types, the relative proportions of different cell types and cell type-specific gene expression profiles in heterogeneous tissues62. Using whole blood, a recent study showed that transcriptional profiles correlate with the extent of tuberculosis disease and change during treatment63. In addition, a specific transcript signature that was characterized by neutrophil-dependent IFNγ and type I IFN signalling could distinguish between active tuberculosis and other inflammatory diseases.
Other recent technological advances may also feature largely in future biomarker research. For example, deep sequencing allows for the detection of epigenetic changes, including alterations in DNA methylation and in the transcription of microRNAs. MicroRNAs do not encode proteins but possess regulatory functions and can alter gene expression. They have been shown to have significant roles in tumour biology, as well as in cardiovascular and rheumatic diseases. MicroRNAs are also involved in regulating inflammation and possibly infectious disease64 and may constitute useful biomarkers for tuberculosis.
Proteomic and metabolomic profiling. Tuberculosis may be differentiated from other infectious and inflammatory conditions based on proteomic fingerprinting of serum using surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) mass spectrometry and ProteinChip array technology65. This single analytical method can detect a very large number of peptides, although it is relatively insensitive. Using this technology, Agranoff and colleagues found that serum amyloid A protein and transthyretin were among the most promising markers to differentiate tuberculosis from other infectious and inflammatory conditions66.
Another approach focuses on the 'metabolome', which is the set of small molecules that encompasses metabolic intermediates, hormones and other signalling molecules, and secondary metabolites67. The potential advantage of metabolomics is the reduced number of potential markers in a single biological specimen. The main disadvantage is that multiple analytical methods seem to be necessary to complete the characterization of these markers. However, no metabolomic data have been reported for tuberculosis to date.
So, unbiased 'omics' approaches, which are largely hypothesis generating and may guide more focused hypothesis-driven approaches, could culminate in qualified biomarkers for use in patient care and in clinical trials.

Conclusions and future perspectives

The challenges posed by M. tuberculosis infection, through its interaction with the immune system and its mechanisms for evasion, require many more breakthroughs from basic science research if we are to make a significant impact on the worldwide tuberculosis problem. Currently, appropriate samples still need to be collected from individuals with clinically characterized protection and susceptibility phenotypes in different populations. The recent advances in technology provide the capacity to search for biomarkers in an unbiased manner using complementary technological platforms, although this requires high-level bioinformatics support.
Host molecules that are present at different levels in clinical phenotypes do not necessarily constitute biomarkers. Most markers discussed in this Review represent differentiating markers, which are markers that are differentially expressed in people with specific outcomes of infection or treatment. These markers are identified on the basis of general exploratory data and have not reached the status of qualified biomarkers.
A very significant effort is required to conduct biomarker validation and biomarker qualification before a differentiating marker can become a qualified biomarker, which is one that has undergone multistep and comprehensive evaluation to confirm precision and accuracy with diagnostic or prognostic value68. New signatures should be validated in a second cohort and may then be used to guide further exploration. This process has to be carefully conducted, as many initially promising markers eventually prove disappointing and many valuable markers are probably never evaluated sufficiently. Our current understanding of host–pathogen interactions and their dynamic nature must be carefully considered in clinical study design to ensure that appropriate and well-differentiated clinical phenotypes are selected for these expensive technologies.
If accurate differentiation between infection and disease states can be achieved, then it could eventually become possible to develop simple point-of-care tests. These could include hand-held devices based on lateral-flow technology that would detect multi-marker signatures (which are much more likely to be successful than single molecules) in patient samples such as serum by immunochromatography (in a similar way to pregnancy tests that detect human chorionic gonadotropin in urine), or hand-held devices for the detection of multi-gene expression signatures.

Competing interests statement

The authors declare no competing financial interests.
top

References

  1. World Health Organisation. Multidrug and extensively drug-resistant TB (M/XDR-TB): 2010 global report on surveillance and response. WHO [online], (2010).
  2. Corbett, E. L. et al. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch. Intern. Med. 163, 1009–1021 (2003).
  3. Abu-Raddad, L. J. et al. Epidemiological benefits of more-effective tuberculosis vaccines, drugs, and diagnostics. Proc. Natl Acad. Sci. USA 106, 13980–13985 (2009).
  4. Davies, P. D. & Pai, M. The diagnosis and misdiagnosis of tuberculosis. Int. J. Tuberc. Lung Dis. 12, 1226–1234 (2008).
  5. Boehme, C. C. et al. Rapid molecular detection of tuberculosis and rifampin resistance. N. Engl. J. Med. 363, 1005–1015 (2010).
    This paper describes the biggest breakthrough in tuberculosis diagnostics in decades: a direct ex vivo M. tuberculosis gene amplification test for the diagnosis and detection of rifampicin resistance.
  6. Black, G. F. et al. BCG-induced increase in interferon-γ response to mycobacterial antigens and efficacy of BCG vaccination in Malawi and the UK: two randomised controlled studies. Lancet 359, 1393–1401 (2002).
  7. Barry, C. E. et al. The spectrum of latent tuberculosis: rethinking the biology and intervention strategies. Nature Rev. Microbiol. 7, 845–855 (2009).
  8. Sudre, P., ten Dam, G. & Kochi, A. Tuberculosis: a global overview of the situation today. Bull. World Health Organ. 70, 149–159 (1992).
  9. Lawn, S. D. & Churchyard, G. Epidemiology of HIV-associated tuberculosis. Curr. Opin. HIV AIDS 4, 325–333 (2009).
  10. Ottenhoff, T. H., Verreck, F. A., Hoeve, M. A. & van deVosse, E. Control of human host immunity to mycobacteria. Tuberculosis 85, 53–64 (2005).
  11. Lawn, S. D., Myer, L., Edwards, D., Bekker, L. G. & Wood, R. Short-term and long-term risk of tuberculosis associated with CD4 cell recovery during antiretroviral therapy in South Africa. AIDS 23, 1717–1725 (2009).
  12. Khader, S. A. et al. IL-23 and IL-17 in the establishment of protective pulmonary CD4+ T cell responses after vaccination and during Mycobacterium tuberculosischallenge. Nature Immunol. 8, 369–377 (2007).
  13. Vordermeier, H. M. et al. Viral booster vaccines improve Mycobacterium bovis BCG-induced protection against bovine tuberculosis. Infect. Immun. 77, 3364–3373 (2009).
  14. Green, A. M. et al. CD4+ regulatory T cells in a cynomolgus macaque model ofMycobacterium tuberculosis infection. J. Infect. Dis. 202, 533–541 (2010).
  15. Lazar-Molnar, E. et al. Programmed death-1 (PD-1)-deficient mice are extraordinarily sensitive to tuberculosis. Proc. Natl Acad. Sci. USA 107, 13402–13407 (2010).
  16. Bruns, H. et al. Anti-TNF immunotherapy reduces CD8+ T cell-mediated antimicrobial activity against Mycobacterium tuberculosis in humans. J. Clin. Invest.119, 1167–1177 (2009).
  17. Stenger, S. et al. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science 282, 121–125 (1998).
  18. Cooper, A. M. Cell-mediated immune responses in tuberculosis. Annu. Rev. Immunol. 27, 393–422 (2009).
  19. Gallegos, A. M., Pamer, E. G. & Glickman, M. S. Delayed protection by ESAT-6-specific effector CD4+ T cells after airborne M. tuberculosis infection. J. Exp. Med.205, 2359–2368 (2008).
  20. Gonzalez-Juarrero, M. et al. Temporal and spatial arrangement of lymphocytes within lung granulomas induced by aerosol infection with Mycobacterium tuberculosis. Infect. Immun. 69, 1722–1728 (2001).
  21. Qin, L., Gilbert, P. B., Corey, L., McElrath, M. J. & Self, S. G. A framework for assessing immunological correlates of protection in vaccine trials. J. Infect. Dis.196, 1304–1312 (2007).
  22. Comstock, G. W. Field trials of tuberculosis vaccines: how could we have done them better? Control. Clin. Trials 15, 247–276 (1994).
  23. Kagina, B. M. et al. Specific T cell frequency and cytokine expression profile do not correlate with protection against tuberculosis after bacillus Calmette-Guerin vaccination of newborns. Am. J. Respir. Crit. Care Med. 182, 1073–1079 (2010).
    This study shows that polyfunctional T cells and cytokine expression do not correlate with BCG-induced protection against tuberculosis.
  24. Darrah, P. A. et al. Multifunctional TH1 cells define a correlate of vaccine-mediated protection against Leishmania major. Nature Med. 13, 843–850 (2007).
  25. Chen, C. Y. et al. A critical role for CD8 T cells in a nonhuman primate model of tuberculosis. PLoS Pathog. 5, e1000392 (2009).
  26. Giri, P. K., Verma, I. & Khuller, G. K. Enhanced immunoprotective potential ofMycobacterium tuberculosis Ag85 complex protein based vaccine against airwayMycobacterium tuberculosis challenge following intranasal administration. FEMS Immunol. Med. Microbiol. 47, 233–241 (2006).
  27. Agger, E. M. et al. Protective immunity to tuberculosis with Ag85B-ESAT-6 in a synthetic cationic adjuvant system IC31. Vaccine 24, 5452–5460 (2006).
  28. Bennekov, T. et al. Alteration of epitope recognition pattern in Ag85B and ESAT-6 has a profound influence on vaccine-induced protection against Mycobacterium tuberculosis. Eur. J. Immunol. 36, 3346–3355 (2006).
  29. Hoft, D. F. et al. Investigation of the relationships between immune-mediated inhibition of mycobacterial growth and other potential surrogate markers of protective Mycobacterium tuberculosis immunity. J. Infect. Dis. 186, 1448–1457 (2002).
  30. Cheon, S. H. et al. Bactericidal activity in whole blood as a potential surrogate marker of immunity after vaccination against tuberculosis. Clin. Diagn. Lab. Immunol. 9, 901–907 (2002).
  31. Spencer, C. T., Abate, G., Blazevic, A. & Hoft, D. F. Only a subset of phosphoantigen-responsive γ9δ2 T cells mediate protective tuberculosis immunity. J. Immunol. 181, 4471–4484 (2008).
  32. Wallis, R. S. et al. Biomarkers and diagnostics for tuberculosis: progress, needs, and translation into practice. Lancet 375, 1920–1937 (2010).
    This is a comprehensive review of diagnostic and other biomarkers for tuberculosis.
  33. Lalvani, A. et al. Enhanced contact tracing and spatial tracking of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Lancet 357, 2017–2021 (2001).
  34. Mori, T. et al. Specific detection of tuberculosis infection: an interferon-γ-based assay using new antigens. Am. J. Respir. Crit. Care Med. 170, 59–64 (2004).
  35. Ferrara, G. et al. Exploring the immune response against Mycobacterium tuberculosis for a better diagnosis of the infection. Arch. Immunol. Ther. Exp. 57, 425–433 (2009).
  36. Pai, M., Zwerling, A. & Menzies, D. Systematic review: T-cell-based assays for the diagnosis of latent tuberculosis infection: an update. Ann. Intern. Med. 149, 177–184 (2008).
  37. Mazurek, G. H. et al. Prospective comparison of the tuberculin skin test and 2 whole-blood interferon-γ release assays in persons with suspected tuberculosis. Clin. Infect. Dis. 45, 837–845 (2007).
  38. Janssens, J. P. Interferon-γ release assay tests to rule out active tuberculosis. Eur. Respir. J. 30, 183–184 (2007).
  39. Losi, M. et al. Use of a T-cell interferon-γ release assay for the diagnosis of tuberculous pleurisy. Eur. Respir. J. 30, 1173–1179 (2007).
  40. Thomas, M. M. et al. Rapid diagnosis of Mycobacterium tuberculosis meningitis by enumeration of cerebrospinal fluid antigen-specific T-cells. Int. J. Tuberc. Lung Dis.12, 651–657 (2008).
  41. Chegou, N. N., Black, G. F., Kidd, M., van Helden, P. D. & Walzl, G. Host markers in QuantiFERON supernatants differentiate active TB from latent TB infection: preliminary report. BMC Pulm. Med. 9, 21 (2009).
  42. Wu, B. et al. Messenger RNA expression of IL-8, FOXP3, and IL-12β differentiates latent tuberculosis infection from disease. J. Immunol. 178, 3688–3694 (2007).
  43. Djoba Siawaya, J. F. et al. Differential cytokine/chemokines and KL-6 profiles in patients with different forms of tuberculosis. Cytokine 47, 132–136 (2009).
  44. Caccamo, N. et al. Multifunctional CD4+ T cells correlate with active Mycobacterium tuberculosis infection. Eur. J. Immunol. 40, 2211–2220 (2010).
  45. Casey, R. et al. Enumeration of functional T-cell subsets by fluorescence-immunospot defines signatures of pathogen burden in tuberculosis. PLoS ONE 5, e15619 (2010).
  46. Harari, A. et al. Dominant TNF-α+Mycobacterium tuberculosis-specific CD4+ T cell responses discriminate between latent infection and active disease. Nature Med. 17, 372–376 (2011).
  47. Steingart, K. R. et al. A systematic review of commercial serological antibody detection tests for the diagnosis of extrapulmonary tuberculosis. Thorax 62, 911–918 (2007).
  48. Steingart, K. R. et al. Performance of purified antigens for serodiagnosis of pulmonary tuberculosis: a meta-analysis. Clin. Vaccine Immunol. 16, 260–276 (2009).
  49. Kunnath-Velayudhan, S. et al. Dynamic antibody responses to the Mycobacterium tuberculosis proteome. Proc. Natl Acad. Sci. USA 107, 14703–14708 (2010).
  50. Hesseling, A. C. et al. Baseline sputum time to detection predicts month two culture conversion and relapse in non-HIV-infected patients. Int. J. Tuberc. Lung Dis. 14, 560–570 (2010).
  51. Johnson, J. L. et al. Shortening treatment in adults with noncavitary tuberculosis and 2-month culture conversion. Am. J. Respir. Crit. Care Med. 180, 558–563 (2009).
  52. Chee, C. B. et al. Tuberculosis treatment effect on T-cell interferon-γ responses toMycobacterium tuberculosis-specific antigens. Eur. Respir. J. 36, 355–361 (2010).
  53. Sai Priya, V. H., Latha, G. S., Hasnain, S. E., Murthy, K. J. & Valluri, V. L. Enhanced T cell responsiveness to Mycobacterium bovis BCG r32-kDa Ag correlates with successful anti-tuberculosis treatment in humans. Cytokine 52, 190–193 (2010).
  54. Wassie, L. et al. Ex vivo cytokine mRNA levels correlate with changing clinical status of Ethiopian TB patients and their contacts over time. PLoS ONE 3, e1522 (2008).
  55. Millington, K. A. et al. Dynamic relationship between IFN-γ and IL-2 profile ofMycobacterium tuberculosis-specific T cells and antigen load. J. Immunol. 178, 5217–5226 (2007).
  56. Millington, K. A., Gooding, S., Hinks, T. S., Reynolds, D. J. & Lalvani, A.Mycobacterium tuberculosis-specific cellular immune profiles suggest bacillary persistence decades after spontaneous cure in untreated tuberculosis. J. Infect. Dis.202, 1685–1689 (2010).
    This paper demonstrates that in some individuals TEM cells persist more than 50 years after spontaneous clinical cure of tuberculosis, suggesting the persistence of antigen. In some cured patients, however, only TCM cells are present, and this is consistent with sterilizing cure.
  57. Bahk, Y. Y. et al. Antigens secreted from Mycobacterium tuberculosis: identification by proteomics approach and test for diagnostic marker. Proteomics 4, 3299–3307 (2004).
  58. Berry, M. P. et al. An interferon-inducible neutrophil-driven blood transcriptional signature in human tuberculosis. Nature 466, 973–977 (2010).
    This work illustrates a role for neutrophil-driven type I IFN signalling in M. tuberculosis pathogenesis and the importance of cell-specific transcriptome analysis.
  59. Jacobsen, M. et al. Candidate biomarkers for discrimination between infection and disease caused by Mycobacterium tuberculosis. J. Mol. Med. 85, 613–621 (2007).
  60. Mistry, R. et al. Gene-expression patterns in whole blood identify subjects at risk for recurrent tuberculosis. J. Infect. Dis. 195, 357–365 (2007).
  61. Maertzdorf, J. et al. Human gene expression profiles of susceptibility and resistance in tuberculosis. Genes Immun. 12, 15–22 (2011).
  62. Repsilber, D. et al. Biomarker discovery in heterogeneous tissue samples – taking the in-silico deconfounding approach. BMC Bioinformatics 11, 27 (2010).
  63. Sartain, M. J., Slayden, R. A., Singh, K. K., Laal, S. & Belisle, J. T. Disease state differentiation and identification of tuberculosis biomarkers via native antigen array profiling. Mol. Cell. Proteomics 5, 2102–2113 (2006).
  64. O'Connell, R. M., Rao, D. S., Chaudhuri, A. A. & Baltimore, D. Physiological and pathological roles for microRNAs in the immune system. Nature Rev. Immunol. 10, 111–122 (2010).
  65. Liu, Q. et al. Serum protein profiling of smear-positive and smear-negative pulmonary tuberculosis using SELDI-TOF mass spectrometry. Lung 188, 15–23 (2010).
  66. Agranoff, D. et al. Identification of diagnostic markers for tuberculosis by proteomic fingerprinting of serum. Lancet 368, 1012–1021 (2006).
  67. de Carvalho, L. P. et al. Activity-based metabolomic profiling of enzymatic function: identification of Rv1248c as a mycobacterial 2-hydroxy-3-oxoadipate synthase.Chem. Biol. 17, 323–332 (2010).
  68. Koulman, A., Lane, G. A., Harrison, S. J. & Volmer, D. A. From differentiating metabolites to biomarkers. Anal. Bioanal. Chem. 394, 663–670 (2009).
  69. World Health Organisation. Treatment of tuberculosis: guidelines 4th edn. WHO[online], (2010).
  70. Diel, R., Loddenkemper, R., Meywald-Walter, K., Niemann, S. & Nienhaus, A.Predictive value of a whole blood IFN-γ assay for the development of active tuberculosis disease after recent infection with Mycobacterium tuberculosis. Am. J. Respir. Crit. Care Med. 177, 1164–1170 (2008).
  71. Doherty, T. M. et al. Immune responses to the Mycobacterium tuberculosis-specific antigen ESAT-6 signal subclinical infection among contacts of tuberculosis patients.J. Clin. Microbiol. 40, 704–706 (2002).
  72. Mattos, A. M. et al. Increased IgG1, IFN-γ, TNF-α and IL-6 responses toMycobacterium tuberculosis antigens in patients with tuberculosis are lower after chemotherapy. Int. Immunol. 22, 775–782 (2010).
  73. Veenstra, H. et al. Changes in the kinetics of intracellular IFN-γ production in TB patients during treatment. Clin. Immunol. 124, 336–344 (2007).
  74. Goletti, D. et al. Is IP-10 an accurate marker for detecting M. tuberculosis-specific response in HIV-infected persons? PLoS ONE 5, e12577 (2010).
  75. Ruhwald, M. et al. Evaluating the potential of IP-10 and MCP-2 as biomarkers for the diagnosis of tuberculosis. Eur. Respir. J. 32, 1607–1615 (2008).
  76. Ruhwald, M., Bjerregaard-Andersen, M., Rabna, P., Eugen-Olsen, J. & Ravn, P. IP-10, MCP-1, MCP-2, MCP-3, and IL-1RA hold promise as biomarkers for infection withM. tuberculosis in a whole blood based T-cell assay. BMC Res. Notes 2, 19 (2009).
  77. Smith, S. G. et al. Mycobacterium tuberculosis PPD-induced immune biomarkers measurable in vitro following BCG vaccination of UK adolescents by multiplex bead array and intracellular cytokine staining. BMC Immunol. 11, 35 (2010).
  78. Demissie, A. et al. Healthy individuals that control a latent infection withMycobacterium tuberculosis express high levels of Th1 cytokines and the IL-4 antagonist IL-4δ2. J. Immunol. 172, 6938–6943 (2004).
  79. Ordway, D. J. et al. Increased Interleukin-4 production by CD8 and γδ T cells in health-care workers is associated with the subsequent development of active tuberculosis. J. Infect. Dis. 190, 756–766 (2004).
  80. Lee, J. H. & Chang, J. H. Changes of plasma interleukin-1 receptor antagonist, interleukin-8 and other serologic markers during chemotherapy in patients with active pulmonary tuberculosis. Korean J. Intern. Med. 18, 138–145 (2003).
  81. Sutherland, J. S., de Jong, B. C., Jeffries, D. J., Adetifa, I. M. & Ota, M. O. Production of TNF-α, IL-12(p40) and IL-17 can discriminate between active TB disease and latent infection in a West African cohort. PLoS ONE 5, e12365 (2010).
  82. Djoba Siawaya, J. F. et al. Differential expression of interleukin-4 (IL-4) and IL-4δ2 mRNA, but not transforming growth factor beta (TGF-β), TGF-βRII, Foxp3, gamma interferon, T-bet, or GATA-3 mRNA, in patients with fast and slow responses to antituberculosis treatment. Clin. Vaccine Immunol. 15, 1165–1170 (2008).
  83. Djoba Siawaya, J. F. et al. Immune parameters as markers of tuberculosis extent of disease and early prediction of anti-tuberculosis chemotherapy response. J. Infect.56, 340–347 (2008).
  84. Eugen-Olsen, J. et al. The serum level of soluble urokinase receptor is elevated in tuberculosis patients and predicts mortality during treatment: a community study from Guinea-Bissau. Int. J. Tuberc. Lung Dis. 6, 686–692 (2002).
  85. Demir, T., Yalcinoz, C., Keskinel, I., Demiroz, F. & Yildirim, N. sICAM-1 as a serum marker in the diagnosis and follow-up of treatment of pulmonary tuberculosis. Int. J. Tuberc. Lung Dis. 6, 155–159 (2002).
  86. Mukae, H. et al. Elevated levels of circulating adhesion molecules in patients with active pulmonary tuberculosis. Respirology 8, 326–331 (2003).
  87. Chan, C. H., Lai, C. K., Leung, J. C., Ho, A. S. & Lai, K. N. Elevated interleukin-2 receptor level in patients with active pulmonary tuberculosis and the changes following anti-tuberculosis chemotherapy. Eur. Respir. J. 8, 70–73 (1995).
  88. Tsao, T. C. et al. Imbalances between tumor necrosis factor-α and its soluble receptor forms, and interleukin-1β and interleukin-1 receptor antagonist in BAL fluid of cavitary pulmonary tuberculosis. Chest 117, 103–109 (2000).
  89. Djoba Siawaya, J. F., Ruhwald, M., Eugen-Olsen, J. & Walzl, G. Correlates for disease progression and prognosis during concurrent HIV/TB infection. Int. J. Infect. Dis. 11, 289–299 (2007).
  90. Rosas-Taraco, A. G. et al. Expression of CDllc in blood monocytes as biomarker for favorable response to antituberculosis treatment. Arch. Med. Res. 40, 128–131 (2009).
  91. Wolday, D. et al. Expression of chemokine receptors CCR5 and CXCR4 on CD4+ T cells and plasma chemokine levels during treatment of active tuberculosis in HIV-1-coinfected patients. J. Acquir. Immune Defic. Syndr. 39, 265–271 (2005).
  92. Hosp, M. et al. Neopterin, β2-microglobulin, and acute phase proteins in HIV-1-seropositive and -seronegative Zambian patients with tuberculosis. Lung 175, 265–275 (1997).
  93. Schleicher, G. K. et al. Procalcitonin and C-reactive protein levels in HIV-positive subjects with tuberculosis and pneumonia. Eur. Respir. J. 25, 688–692 (2005).
  94. Krenke, R. & Korczynski, P. Use of pleural fluid levels of adenosine deaminase and interferon γ in the diagnosis of tuberculous pleuritis. Curr. Opin. Pulm. Med. 16, 367–375 (2010).
  95. Abel, B. et al. The novel tuberculosis vaccine, AERAS-402, induces robust and polyfunctional CD4+ and CD8+ T cells in adults. Am. J. Respir. Crit. Care Med. 181, 1407–1417 (2010).
  96. Hohn, H. et al. MHC class II tetramer guided detection of Mycobacterium tuberculosis-specific CD4+ T cells in peripheral blood from patients with pulmonary tuberculosis. Scand. J. Immunol. 65, 467–478 (2007).
  97. Axelsson-Robertson, R. et al. Extensive major histocompatibility complex class I binding promiscuity for Mycobacterium tuberculosis TB10.4 peptides and immune dominance of human leucocyte antigen (HLA)-B*0702 and HLA-B*0801 alleles in TB10.4 CD8 T-cell responses. Immunology 129, 496–505 (2010).
  98. Veenstra, H. et al. Changes in leucocyte and lymphocyte subsets during tuberculosis treatment; prominence of CD3dimCD56+ natural killer T cells in fast treatment responders. Clin. Exp. Immunol. 145, 252–260 (2006).
  99. Azzurri, A. et al. Serological markers of pulmonary tuberculosis and of response to anti-tuberculosis treatment in a patient population in Guinea. Int. J. Immunopathol. Pharmacol. 19, 199–208 (2006).
  100. Esquivel-Valerio, J. A. et al. Antineutrophil cytoplasm autoantibodies in patients with tuberculosis are directed against bactericidal/permeability increasing protein and are detected after treatment initiation. Clin. Exp. Rheumatol. 28, 35–39 (2010).
  101. Jacobsen, M. et al. Suppressor of cytokine signaling (SOCS)-3 is affected in T cells from TB patients. Clin. Microbiol. Infect. 29 Jul 2010 (doi: 10.1111/j.1469-0691.2010.03326.x).
top

Author affiliations

  1. Immunology Research Group, NRF/DST Centre of Excellence for TB Biomedical Research, Division of Molecular Biology and Human Genetics, Faculty of Health Sciences, Stellenbosch University, South Africa.
  2. South African Tuberculosis Vaccine Initiative, Institute of Infectious Diseases and School of Child and Adolescent Health, University of Cape Town, South Africa.
  3. Division of Infection and Immunity, University College London Medical School, London, UK.
Correspondence to: Gerhard Walzl1 Email: gwalzl@sun.ac.za

Postado por às
Marcadores:

Nenhum comentário:

Postar um comentário

Assinar: Postar comentários (Atom)