**2. Blood tests in TB diagnosis**

**1. Introduction**

2 Basic Biology and Applications of Actinobacteria

consequently control plan.

diagnostic setup [6].

Tuberculosis caused by *M*. *tuberculosis* is the leading cause of human deaths from single infectious agent and fall in the top 10 causes of deaths worldwide. More than 10 million people suffered from TB in 2016 and the global case fatality and incident rates were 16% and 140/100,000 [1].

Inhalation of *M*. *tuberculosis* stimulates host cellular immune response with a consequence of either clearance of the organism or infection. Infection may result into latent mycobacteria infection (LTBI) (no clinical evidence) [2] or active tuberculosis (pulmonary or extra-pulmonary) [1].

The global LTBI prevalence is 23%, which is approximately 1.7 billion of the population [3]. About 5–15% of LTBI may progress to active TB [4]. LTBI is currently diagnosed by tuberculin skin test (TST) or gamma-interferon release assays (IGRA) [5]. On average the prevalence of

The zoonotic potential of *M. bovis* constitutes a public health concern. Ingestion of *M. bovis* contaminated material is reported to be a primary cause of infection in humans and the resulting TB may take both pulmonary and extra-pulmonary form [6]. The burden of tuberculosis in humans due to *M. bovis* is variable in different countries, ranging from 5 to 28% with mortalities of 8.7% [6]. Therefore, this pathogen should be given due weight in TB diagnosis and

According to WHO report, 53 million deaths have been averted from 2000 to 2016 following diagnosis and treatment. There is a big gap between diagnosis and treatment otherwise the number of deaths prevented could be even higher [1]. Every year the health systems fail to capture 3 million TB patients [7] some of them may be due to the nature of mycobacteria and

When prevention of pathogen-host contact is inadequate, early, rapid, and discriminative diagnosis become the primary factor for treatment success and reduction of further transmissions. Inappropriate diagnosis of symptomatic patients with active TB, LTBI, or other confounding diseases, for example, sarcoidosis, usually lead to delayed or wrong administration of full TB treatment or prophylactic dose [8]. This chapter, therefore, focuses on mycobacterial infections and tuberculosis diagnosis by assessing and comparing different procedures so as

The assessment and comparison of diagnostic procedures in this chapter are based on either single or combination of features such as the type of sample tested, pathological, or morphological changes or immunological reactions. Some of the known mycobacterial infections and tuberculosis diagnostic sample sources include sputum for smear, sniffing, and culture, blood for biomarkers such as interferon, inflammatory cells, proteins and transcriptomes, urine, and breath for volatile organic compounds (VOCs). Moreover, anatomical and physiological changes in tissues can allow diagnosis of tuberculosis through imaging techniques and skin reaction. Among these procedures, IGRA, TST and transcriptomic studies can detect mycobacterial infections while smear microscopy, culture and sniffing results of sputum,

to suggest a best diagnostic method(s) and research direction.

active TB globally is 0.14% and the figures may vary according to regions [1].

Whole blood sample can be used to diagnose both mycobacterial infections and tuberculosis based on the host immune response. Characteristics of blood parameters such as interferon-induced genes, myeloid genes, inflammatory genes and B and T-cells regulation genes, and proteomic profiles are used as markers in the detection of mycobacterial infections and tuberculosis [9].

#### **2.1. Interferon-γ release assays (IGRAs)**

IGRAs are immune cell-mediated in vitro blood tests that measure mononuclear cell (lymphocytes: T and B cells) release of interferon-γ (INF-γ) after stimulation by antigens specific for Mycobacterium. Following infection, the body immune response is triggered and immune cells such as macrophages, T-cells, B-cells, and natural killer cell are involved. Macrophages, which form the first defense line, engulf, kill, and eliminate *Mycobacteria tuberculosis*. Some *M*. *tuberculosis* escapes the immune mechanism and survives and replicate in macrophages. T-cells, especially CD4 and CD8 T-cells, produce cytokine—interferon-gamma, which in turn activate infected macrophages to produce reactive nitric oxide and related reactive nitrogen intermediates to kill the *M*. *tuberculosis* and eliminate it through the actions of phagosomes and lysosomes [10]. The lymphocytes can produce interferon-gamma in vitro whenever stimulated by *M*. *tuberculosis* antigen. This feature is used to assist in the diagnosis of *Mycobacteria tuberculosis* infection. Principally, the lymphocytes of a person infected with *M*. *tuberculosis* release interferon-gamma when mixed with *M*. *tuberculosis* derived antigens. There are two enzyme-linked assays; enzyme-linked immunosorbent assay (ELISA), and enzyme-linked immunosorbent spot (ELISPOT) assay. The assays detect INF-γ release by lymphocytes of sample donor following exposure to antigens found on the *M*. *tuberculosis* complex (MTBC).

In ELISA based IGRA, fresh whole test blood sample is mixed with *M. tuberculosis* antigens and control. The antigen is made up of 6kDaA early secreted antigen target (ESAT-6), 10-kDa culture filtrate protein (CFP-10), and TB7.7 [11]. Normal saline is used as a control. The assay gives the concentration of INF-γ in international units per milliliter (IU/ml). The test result is considered positive when the difference in INF-γ response to the TB antigen between the test antigen value and the control value is greater than the cut-off of 0.35 IU [12].

Principally, the ELISPOT assay works on peripheral blood mononuclear cells (PBMCs) from a whole blood sample. The PBMCs are mixed with synthetic peptide antigens (ESAT-6 and CFP-10) and the control and incubated to stimulate interferon-γ secretion by the cells. Secreted interferon-γ is captured by specific antibodies and accumulated as spots. Thus, ELISPOT assay gives the number of interferon-γ producing cells (spots). The test result is considered positive when the difference in a number of spots between the test sample and the negative control is greater than eight (8) [13].

statistically analyzed to define mass spectral patterns. Results show unique abundant spectra (up-regulation) in plasma from subjects with active TB, which are distinct from those of sub-

Paralleling of Diagnostic Endeavor for Control of Mycobacterial Infections and Tuberculosis

http://dx.doi.org/10.5772/intechopen.77316

5

Proteomic fingerprinting of plasma by surface-enhanced laser desorption/ionization—time of flight (SELDI-TOF) demonstrates that active TB can be discriminated from LTBI and other confounding diseases. Sandhu and colleagues [8] have reported that plasma from active TB patients had more distinct spectral peaks (at 5.8–11.5 kDa) than symptomatic LTBI and non-LTBI controls. The procedure discriminates the active TB patients from the undifferentiated controls with an accuracy of 87% (sensitivity 84% specificity 90%). Moreover, active TB can be differentiated from symptomatic controls with LTBI at 87% accuracy (sensitivity 89%, specificity 82%) and from symptomatic controls without LTBI at 90% accuracy (sensitivity 90%, specificity 92%). In this study Serum Amyloid A and transthyretin were reported as potential protein biomarkers for TB diagnosis. Similar procedure was also used by Agranoff and colleagues [18] on a serum to distinguish active TB patients from symptomatic controls, with an accuracy of 94% (sensitivity 93·5%, specificity 94·9%). Again, Serum Amyloid A and

Liu and colleagues have also used SELDI-TOF MS to screen serum samples from TB patients and controls (lung cancer, pneumonia, chronic obstructive pulmonary disease, and healthy volunteers). Among different TB related peaks that were detected, fibrinogen with mass to charge ratio of 2554.6 Da was up-regulated in TB patients compared to controls. This discrimination of TB patients from controls had an accuracy of 83.8% (sensitivity of 83.3% and specificity of 84.2%) and suggests that fibrinogen could be a potential TB biomarker [19]. A similar procedure was performed by Zhang and colleagues [20], whereby TB patients were differentiated from controls (non-TB controls such as pulmonary cancer, pneumonia, chronic bronchitis, emphysema, asthma and flu, and healthy subjects) with a sensitivity of 96.9% and specificity of 97.8% (accuracy up to 97.3%). The most distinct protein peak at mass/charge ratio of 5643 Da was up-regulated in TB patients and identified as orosomucoid protein.

In most proteomic finger printings for TB diagnosis, comparison of active TB patients, LTBI, and controls (healthy subjects or patients with other non-TB diseases) has been done. The peaks of identified protein biomarkers in controls have served as references. However, the

According to expert opinions, blood-based mycobacterial diagnostic procedures for example, IGRAs can give supporting information in diagnosis in some situations like extra-pulmonary TB, testing of negative acid-fast bacilli (AFB) in sputum and/or culture negative for *M. tuberculosis*, TB diagnosis in children, or in the differential diagnosis of infection with non-tuberculous mycobacteria (NTM) [21]. IGRA can also support the diagnosis of *M. bovis* infection when

Sputum of patients with pulmonary TB contains mycobacterium pathogens, which serves a role in transmission. The presence of the pathogens and their features such as viability,

transthyretin were identified as potential biomarkers for TB diagnosis.

single intradermal comparative tuberculin test (SICTT) is negative [22].

identified protein biomarkers are not TB specific.

**3. Sputum tests in TB diagnosis**

jects with LTBI or controls.

Both IGRAs have the ability to diagnose latent mycobacterial infection in 24 h following the one-time patient visit and submission of the blood sample to a health center [11]. The sensitivity of the IGRAs in adults is 80–90% when mycobacterium standard culture is used as gold [13]. In contrast to tuberculin skin test, IGRAs do not cross-react with Bacille Calmette-Guérin (BCG) antigen hence prior BCG vaccination does not cause false positives [11]. IGRAs cannot discriminate active tuberculosis from LTBI and its sensitivity is low in immunocompromised patients and children under the age of 5 years [13]. The cost of mycobacterium infection diagnosis by interferon- γ release assays is more than 40 USD per sample, which is higher compared to 10.56–25.97 USD for tuberculin skin test [14, 15]. This factor may influence the availability of and accessibility to this diagnostic service, especially in low-income countries.

#### **2.2. Mycobacterium global gene signature/transcriptional blood signatures**

Whole blood transcriptional signature can assist in distinguishing latent from active tuberculosis by showing features, which are not present in asymptomatic patients. Moreover, the transition from latent infection to active TB can be predicted by assessing the blood parameter [9]. This approach is important because of the reported 10–20% of detectable tuberculosis in asymptomatic individuals diagnosed with latent *M. tuberculosis* infection [16].

TB risk signature or global gene expression based on whole blood RNA genotyping polymerase chain reaction (PCR and sequencing) can be used to predict the chance of progression from latent to active TB. It has been reported that the expression of signature genes (that is quantities of functional RNA) increases toward active TB development in contrast to non-TB progressor (latent mycobacterium infection). For instance, Zak and colleagues [17] have reported a 16-gene signature with a sensitivity of 71.2% and specificity of 80% (95% CI: 66·6–75·2) at a 6 month time before tuberculosis diagnosis. This approach shows promising future of TB diagnosis as early as possible when samples are submitted since PCR and sequencing can give diagnostic results within 24 h.

Transcriptional blood signatures studies have been conducted in many parts of the world including high-income countries (US, UK, and German), middle-income countries (Brazil, South Africa, and Indonesia) and low-income countries (Kenya, Gambia, and Malawi) [9]. However, the procedure is under research and there is hope that it will become a handy tool in TB diagnosis and treatment decision.

#### **2.3. Blood proteomic profiling/fingerprinting**

Proteome, the entire set of proteins produced or modified by a cell, system or organism, vary with time, requirement, stress or disease state that cell, and system or organism experiences. Studies on types and levels of plasma or serum proteins show variations between individuals with LTBI, active TB and those under different stages of treatment. These features can be used to differentiate LTBI from active TB and other perplexing diseases. Plasma or serum protein peaks are identified using high-performance liquid chromatography (HPLC)-tandem matrix-assisted laser desorption/ionization-TOF-MS (MALDI-TOF-MS) and the results are statistically analyzed to define mass spectral patterns. Results show unique abundant spectra (up-regulation) in plasma from subjects with active TB, which are distinct from those of subjects with LTBI or controls.

Proteomic fingerprinting of plasma by surface-enhanced laser desorption/ionization—time of flight (SELDI-TOF) demonstrates that active TB can be discriminated from LTBI and other confounding diseases. Sandhu and colleagues [8] have reported that plasma from active TB patients had more distinct spectral peaks (at 5.8–11.5 kDa) than symptomatic LTBI and non-LTBI controls. The procedure discriminates the active TB patients from the undifferentiated controls with an accuracy of 87% (sensitivity 84% specificity 90%). Moreover, active TB can be differentiated from symptomatic controls with LTBI at 87% accuracy (sensitivity 89%, specificity 82%) and from symptomatic controls without LTBI at 90% accuracy (sensitivity 90%, specificity 92%). In this study Serum Amyloid A and transthyretin were reported as potential protein biomarkers for TB diagnosis. Similar procedure was also used by Agranoff and colleagues [18] on a serum to distinguish active TB patients from symptomatic controls, with an accuracy of 94% (sensitivity 93·5%, specificity 94·9%). Again, Serum Amyloid A and transthyretin were identified as potential biomarkers for TB diagnosis.

Liu and colleagues have also used SELDI-TOF MS to screen serum samples from TB patients and controls (lung cancer, pneumonia, chronic obstructive pulmonary disease, and healthy volunteers). Among different TB related peaks that were detected, fibrinogen with mass to charge ratio of 2554.6 Da was up-regulated in TB patients compared to controls. This discrimination of TB patients from controls had an accuracy of 83.8% (sensitivity of 83.3% and specificity of 84.2%) and suggests that fibrinogen could be a potential TB biomarker [19]. A similar procedure was performed by Zhang and colleagues [20], whereby TB patients were differentiated from controls (non-TB controls such as pulmonary cancer, pneumonia, chronic bronchitis, emphysema, asthma and flu, and healthy subjects) with a sensitivity of 96.9% and specificity of 97.8% (accuracy up to 97.3%). The most distinct protein peak at mass/charge ratio of 5643 Da was up-regulated in TB patients and identified as orosomucoid protein.

In most proteomic finger printings for TB diagnosis, comparison of active TB patients, LTBI, and controls (healthy subjects or patients with other non-TB diseases) has been done. The peaks of identified protein biomarkers in controls have served as references. However, the identified protein biomarkers are not TB specific.

According to expert opinions, blood-based mycobacterial diagnostic procedures for example, IGRAs can give supporting information in diagnosis in some situations like extra-pulmonary TB, testing of negative acid-fast bacilli (AFB) in sputum and/or culture negative for *M. tuberculosis*, TB diagnosis in children, or in the differential diagnosis of infection with non-tuberculous mycobacteria (NTM) [21]. IGRA can also support the diagnosis of *M. bovis* infection when single intradermal comparative tuberculin test (SICTT) is negative [22].
