Infection and the host immune system have an intimate relationship as two facets of the same coin. When Mtb enters the host lung via inhalation and invades interstitial tissue, the immune response is inevitably activated, resulting in host profile alterations (Bulterys et al., 2019; Yong et al., 2019). Since they are easy to use, affordable, and practical for POC, pathogen-specific antibodies are most often utilized as the host biomarkers for serologic tests. The Mtb-specific antibodies have been extensively investigated for infection diagnosis and response prediction to anti-TB treatment, including IgA, IgG, and IgM antibodies to: mycobacterial secreted 38-kDa and cytosolic 16-kDa antigens (Raja et al., 2008; Ben-Selma et al., 2011); antibodies against culture filtrate protein 10 (CFP-10) and early secretory antigenic target 6 (ESAT-6) (Andersen et al., 2000; Arend et al., 2000); and antigen 60, heat shock protein, purified protein derivative, and lipid-derived antigens (Kumar Verma and Jain, 2007). Even though participants had smear-positive results, earlier investigations revealed a low sensitivity diagnosis, approximately 75% at best, by ELISA (Raja et al., 2008). The heterogeneous quality and quantity of host antibody responses could be revealed as the main reason, and probably arise at different stages of the disease or are regulated by multiple host factors. Recently, several promising candidates were reported to improve the sensitivity and specificity of TB diagnosing assays, some widely common, such as: serum IgG against lipoarabinomannan (LAM) (Ben Selma et al., 2010); RV1255c-E and RV0310c-E antigens (Luo et al., 2017); proline-proline-glutamic acid protein 17 (Abraham et al., 2018); and mycobacterial DNA binding protein 1 (Maekura et al., 2019).

Host immunity protection against Mtb is not fully understood, but it depends on various innate and adaptive immunological mechanisms. Additionally, cellular immune responses to Mtb-specific antigens have demonstrated higher constancy than antibody responses (Yong et al., 2019). In principle, T cell-induced immune responses play a crucial role in controlling and protecting against Mtb infection (Walzl et al., 2011). At certain differentiated stages, CD4⁺ and CD8⁺ T cells progress to central memory (T_CM) and effector memory (T_EM), which activate and modulate phagocytosis and produce molecules that can directly influence cytotoxic microbial infection. Based on these understandings, several attempts have been undertaken to characterize the T cell-functional signature correlated with different TB stages. In particular, CD69 or CD137, a co-stimulatory receptor, is responsible for T cell activation and proliferation, and the frequency of CD69⁺ and CD137⁺ has been reported to link with active TB (Nikolova et al., 2013; Yan et al., 2017). Another study by Ahmed et al. used two markers for immune activation, CD38, and maturation, CD27, on Mtb-specific CD4⁺ T cells to differentiate active TB and LTBI (Ahmed et al., 2018). According to their observations, CD38^pos CD27^low frequencies were increased in active TB whilst LTBI performed CD38^neg CD27^high profile.

Treatment efficacy can be monitored via promising surface markers such as T-cell activation (CD38, HLA-DR) and proliferation (Ki-67) (Adekambi et al., 2015) under phenotypic characteristic analysis on Mtb-specific T cells. After nine weeks of anti-TB therapy initiation, CD38, Ki67, and HLA-DR frequencies of T cells were remarkably decreased and associated with stable culture conversion time (Ahmed et al., 2018). The amount of T_EM and T_CM could also be exerted as a predictor for reoccurrence, in which after six months of anti-TB therapy initiation the cases with a substantial amount of T_EM may result in relapse because of living Mtb persistence, while T_CM may reflect the complete clearance of Mtb (Millington et al., 2010; Wang et al., 2010).

Various host cytokines and chemokines are released once Mtb-specific antigens encounter and stimulate the immune responses of an Mtb-infected individual (Baatjies et al., 2021). The intricate relationship between Mtb and the immune system is attributed to the production of various cytokines in response to Mtb infection (Etna et al., 2014) which could restrict bacterial growth and modulate inflammatory responses that contribute to the development of TB (Bai et al., 2018). Additionally, chemokines are supposed to be closely associated with the development and maintenance of quiescent granulomas as well as enrolling periphery cells to positioning within the granuloma (Kumar et al., 2019b). Diverse investigations have focused on cytokine and chemokine identification which can improve discrimination between LTBI and active TB, treatment monitoring, and outcome prediction ( Table 1 ).

The most popular commercial immune-based diagnosis for TB in clinical practice is still IGRAs despite the fact that IGRAs do not discriminate between active TB and LTBI and are perhaps inefficient in high-burden conditions (Organization, 2011; Pai et al., 2014; Jacobs et al., 2016). Hence, numerous studies have been reported to identify several singleplex or complex accurate and sensitive cytokine-based biomarkers for serological tests. In particular, cytokine/chemokine-based biomarkers can be used for monitoring treatment response in TB patients. Another examination was performed with plasma chemokines of PTB, LTBI, and healthy control (HC) individuals (Kumar et al., 2019b) to determine mycobacterial burden and disease severity. This study exhibited that the PTB chemokine levels containing CCL1, CCL3, CXCL1, CXCL2, CXCL9, and CXCL10 are significantly higher compared to LTBI and HC individuals. Furthermore, Kumar et al. also found that the plasma levels of CCL1, CCL2, CXCL2, CXCL9, CXCL10, and CXCL11 are all remarkably reduced following anti-TB chemotherapy, suggesting that they can be added to host chemokine panels for treatment-response monitoring. A meta-analysis was performed to identify valuable cytokines for distinguishing LTBI and active TB (Wei et al., 2020). Fourteen studies with 982 participants evaluated AUC, sensitivity, and specificity to reveal potential immune markers. As a result, seven molecules were identified that can discriminate between individuals with LTBI and active TB, including IL-2 (AUC, 0.91; sensitivity, 0.87; specificity, 0.61), IP-10 (0.86, 0.77, and 0.73), IL-5 (0.85, 0.64, and 0.75), IL-13 (0.85, 0.75, and 0.71), IFN-γ (0.80, 0.67, and 0.75), IL-10 (0.80, 0.68, and 0.74), and TNF-α (0.78, 0.67, and 0.64). Notably, due to their low specificity, single cytokines barely exhibit adequate diagnostic performance to be used as biomarkers. Also, host immune signatures have been developed in primary care settings (Mutavhatsindi et al., 2021). However, there is insufficient evidence to support their validity in secondary care settings.

Primarily evidence revealed that the host immune response to Mtb antigens is suppressed in resistant TB, but inflammatory cytokines still serve a fundamental role in the persistence of pulmonary tissue damage (Fortes et al., 2005). Impaired Th1-related cytokine production (Lee et al., 2002; Fortes et al., 2005; Geffner et al., 2009), such as IFN-γ, and increased Th2-related cytokines (Geffner et al., 2009), such as IL-4, were observed early in blood samples collected from MDR- and drug susceptible TB (DS-TB) patients, which emphasizes the alteration of Th1/Th2 profile characteristics in the disease progression. Furthermore, at the initiation of treatment for MDR-TB patients, peripheral Th1 and Th2 cells are remarkably diminished in comparison with drug susceptible TB (DS-TB) and HCs. These observations suggested the inhibition of both Th1 and Th2 pathways were intervened in the host response to MDR-TB infection (Tan et al., 2012). Besides the decreased level of IFN-γ, it was also found that IL-2 cytokines are significantly diminished in MDR-TB blood, even though they can function as protective cytokines against MDR-TB infection. Notably, different cytokine profiles are characterized in each stage of MDR-TB patients, with significantly decreased levels of IFN-γ, IL-2, and IL-10 in early infection, while during progressive necrosis and fibrosis, IL-4, IL-6, and TNF-α are increased (Tan et al., 2012). These results show that MDR-TB infection is undoubtedly complicated and challenging to treat. Thus, it is necessary to undertake extensive examinations for MDR-TB cytokine profiles that can identify biomarkers for MDR-TB diagnosis and treatment monitoring.

There are fluctuating cytokine profiles that coincide during the treatment process from the transition between intensive TB treatment and the continuation phase (Riou et al., 2012). The data from the study of Riou and colleagues demonstrated that only plasma levels of IP-10 and vascular endothelial growth factor (VEGF) remarkably changed in response to anti-TB treatment (Riou et al., 2012). These results substantiated previous observations that the reduction of plasma IP-10 is associated with successful TB treatment (Azzurri et al., 2005; Djoba Siawaya et al., 2009); hence IP-10 could be utilized as a valuable indicator of clinical response and treatment success. Furthermore, an investigation was recently conducted that quantified serum inflammatory and anti-inflammatory cytokine concentrations and compared MDR-TB, DS-TB, and HC (Mensah et al., 2021). The results identified a low concentration of serum IFN-γ and a high level of IL-4 related to MDR-TB compared to DS-TB individuals. Moreover, in the measurement of proinflammatory and anti-inflammatory cytokines, Granzyme B, IFN-γ, IL-4, IL-6, and TNF-α showed the greatest promise to differentiate between DS-TB or MDR-TB and HC. A recent systemic review identified promising biomarkers for the effective assessment of TB therapy, indicating that TNF-α is the most valuable biomarker (Clifford et al., 2015). Moreover, the proportion of both IFN-γ and IL-2 was reported to increase during the treatment period, suggesting they may be common markers of immunological protection against TB. However, the cytokine profiles from reported studies that focused on treatment monitoring in TB are controversial due to the heterogeneity of study designs and small sample sizes (Clifford et al., 2015), so extensive investigation is required to predict any failure of first-line treatments and ensure the adequacy of LTBI.

Despite treatments initially seeming to be effective, TB relapse occasionally occurs after completing anti-TB therapy and clinical symptoms re-emerge (De Steenwinkel et al., 2013). Transition to a negative sputum culture is generally considered predictive for effective TB treatment after two months (Wallis et al., 2013). However, this predictive method has poor sensitivity, with only 40% predicting relapse and 57% predicting failure outcome (Horne et al., 2010). Recently, emerging studies indicated that the failures of the host immune system to restrain invading Mtb, including immune disorders or severe immunodeficiency, are supposed to be the leading cause of re-treated TB (Bai et al., 2018). Bai et al. observed that cytokines released from Th1 and Th2 cells dramatically altered and upregulated Th1/Th2 cytokine ratios in re-treated TB patients (Bai et al., 2018). Several serum markers, including IFN-γ, IL-2, IL-7, and soluble CD54, were shown remarkably enhanced and more sensitive compared with other cytokines. Hence, they might serve as helpful serum indicators to determine TB re-treatment response. Moreover, a study from Steenwinkel et al. compared plasma cytokine profiles of primary TB and re-occurrence TB, especially in patients who were treatment non-compliant (De Steenwinkel et al., 2013). Not only the Mtb load in infected organs, additionally, in the cases of HIV-infected individuals or patients suffering a long period of immunosuppressive therapy with TNF-α inhibitors, the reactivated risk of LTBI increased remarkably (Bruns et al., 2009; Getahun et al., 2010; Amelio et al., 2019). Hence, besides biomarkers that imply successful treatment during the therapeutic period, more specific indicators are required to predict relapse at the end of six-month treatment, which might significantly enhance clinical prognosis. Interestingly, an investigation from Ronacher et al. disclosed that the levels of IL-5 and MMP-2 decreased, while sIL-2Rα and CRP were highly elevated at diagnosis in relapse patients in comparison with failed and cured individuals (Ronacher et al., 2019).

An immunomics approach has been widely investigated to differentiate between active TB, LTBI, and/or healthy people (Wei et al., 2020; Mutavhatsindi et al., 2021). Earlier publications found higher proportions of neutrophils and monocytes in active TB compared to non-active TB patients (Wang et al., 2015; La Manna et al., 2017; Estévez et al., 2020a). On the other hand, the proportions of T-helper (CD4⁺) cells and B cells were higher in non-active TB, suggesting that adaptive immune response is lower in active TB patients. However, as it appears more challenging, the immune expression profiles of those with LTBI largely overlapped with HC, making it difficult to distinguish separate clusters based on the 1,000 most variable genes (Bah et al., 2018). It was discovered that the type I and type II interferon signaling pathways’ upstream and downstream components are expressed more strongly in peripheral blood leukocytes from active TB patients (Berry et al., 2010; Zambuzi et al., 2016). Additionally, another study has shown that the expression of CD64, LTF, and Rab33A together formed biomarkers that can distinguish individuals with and without TB (Jacobsen et al., 2007). Furthermore, it has been widely recognized that IL-6 could be useful for monitoring the effectiveness of anti-TB drugs and, together with sCD14, used to distinguish the active state of the disease (Yong et al., 2019; Boni et al., 2022). Previous reports also showed that in active TB patients, plasma levels of sCD163 and TNF-α were linked with the severity of the condition and extent of lung damage (Zambuzi et al., 2016; De Andrade Júnior et al., 2008; De Steenwinkel et al., 2013). Another study also identified several increased immune markers were linked to active TB patients, reported as GM-CSF, IFN-α-2, IL-4, IL-5, IP-10, and MIP-1α (Zambuzi et al., 2016; Estévez et al., 2020b). Until now, the diagnosis confirmation of LTBI patients still followed the results of the IGRA test (Herrera et al., 2011). Nevertheless, several limitations, such as the large amount of blood needed, limited assay reproducibility, low sensitivity, unknown prognostic value, and slow sample processing, have limited IGRA usage for mass testing (Herrera et al., 2011; Kaul et al., 2022). A recent publication confirmed FCGR1B, GBP1, and GBP5 as the transcript signatures that distinguished between LTB1 and active TB, which could be helpful in screening LTBI on a mass scale (Kaul et al., 2022). Furthermore, they classified the LTBI into clusters following the IFN-γ values in TB1 and TB2 antigens using machine-learning cluster analysis. Such a clustering approach may help to identify which group would likely progress to become active TB, thus effectively targeting preventive therapy in this cluster. However, the implementation of immunomics to distinguish between active TB and LTBI remains challenging. Even though these biomarkers could be further used to monitor treatment, differentially diagnose, and predict disease progression, little to none of these markers have been validated in extensive prospective studies. The ratio among biomarkers might keep changing inconstantly throughout the treatment phase and be considered as a dynamic of biomarkers, another critical challenge to be addressed (Dheda et al., 2010). Therefore, utilization of homologous (e.g., gene experiment only) or heterogenous (e.g., cytokines and endogenous metabolites) biosignatures to distinguish disease diagnoses and treatment monitoring would be much more reliable and preferable, rather than the identification of single biomarkers.

Special or edge populations usually appear with physiological changes or system immaturities that may produce different immune cell markers (Ciabattini et al., 2018; Abu-Raya et al., 2020; Aulin et al., 2021). Pediatric, pregnant, elderly, and immunocompromised patients are recognized as populations that confound standard TB treatment strategies (Nahid et al., 2016). Determining the TB diagnosis in the pediatric population remains challenging with current routine clinical and laboratory tests (Gunasekera et al., 2022). Both the TST and IFN-γ fail to differentiate diagnoses of TB in pediatric patients. Hence, the WHO have promoted the pediatric TB scoring chart that was invented to help with assessment in limited-resource countries which might be biased (Van Rheenen and Health, 2002). Furthermore, treatment monitoring remains elusive in this population. Most of the time, a clinician only assesses successful treatment based on weight gain and loss of symptoms (Chiang et al., 2020). Although an immune markers-based test promises to solve this issue, no immune-diagnostic or monitoring test is well-developed and validated (Yong et al., 2019; Kumar et al., 2021a). A recent publication found that pediatric TB patients have 20% differentially expressed genes in comparison to adults (Bah et al., 2018). It is very well-known that childhood patients are characterized notably by the significant increase of IFN-regulated genes (Yoshida et al., 2022). Nonetheless, other regulated immune pathway genes may also be upregulated, including the elevation of alternative antimicrobial defense mechanisms (H2AFJ, HISTIH2BG, and CTSG) and multi-functional IL-27 that blocks IL-17 and IRF1 signaling (Bah et al., 2018). In addition to upregulated genes, pediatric TB is also characterized by downregulated genes, with pathways primarily engaged in stimulation of T or B cells (CD40L, CD7, ICOS, FCER2, PTPRCAP, and ADAM23), dendritic cell development (FLT3LG), alternative promoters of inflammation (CD248 and EDAR), and inhibition of neutrophil degranulation by ADORA3.

Due to their central role in promoting immune responses, monocytes and lymphocytes may reflect a person’s immune status during infection. In TB infection, Mtb will target the monocytes for their growth, while lymphocytes are principally responsible for Mtb clearance. Monocyte to lymphocyte ratio (ML ratio) has gained interest as a potential biomarker that can classify TB diagnosis even in special populations. A previous report investigating infants born to HIV-positive mothers revealed that the increased ML ratio in peripheral blood was associated with definite TB disease. Although the ML ratio could be a good predictor for TB disease, the ratio showed a modest role in predicting TB disease-free survival (Naranbhai et al., 2014). Another study reported that an extreme deviation of the ML ratio, either high or low (median of ML ratio: 0.36), was correlated with active TB disease. The authors also mentioned that patients with EP-TB and an age of >60 years were more likely to have severe ML ratio alterations (Wang et al., 2015). Since Mtb infection can interfere with the development of hematopoietic stem cells (HSCs), the ML ratio might be affected differently among people. For example, Mtb hinders HSCs engraftment and uniquely reorganizes HSCs (Khan et al., 2020); Mtb infection on HSCs could wreak havoc on the balance of myeloid and lymphoid lineages development in different people (Tamburini et al., 2021). This condition could return to normal with the administration of anti-TB therapy. On the contrary, an earlier study failed to distinguish between LTBI and active TB using the ML ratio (Buttle et al., 2021). Even though the ML ratio showed great sensitivity, the specificity was poor. However, this study discovered a higher ML ratio in males due to less effective immune responses in males. Therefore, the severity of TB disease might be more advanced in male patients. Another study confirmed this finding, showing that the ML ratio (median: 0.5) could distinguish patients with active TB from those of healthy volunteers (La Manna et al., 2017). Despite this, it failed to distinguish active TB from LTBI patients and cured TB patients. Apart from the ML ratio, the neutrophil to lymphocyte ratio (NL ratio) has recently been proposed as a biomarker to distinguish active TB, LTBI, cured TB, and HC. A recent report found that active TB patients appeared to have a high NL ratio compared to both HC and cured TB patients (La Manna et al., 2019). Among those with active TB, a higher neutrophil count and a lower lymphocyte count were identified. Nevertheless, this ratio was gradually decreased by TB treatment, showing the potential of this marker to monitor TB as well. Understanding these potential biomarkers to diagnose TB is considerably important to provide faster and easier confirmation of TB. Therefore, anti-TB drugs can be started immediately. The aforementioned biomarkers were routinely performed in hospital laboratory settings, even in low-resource countries. Although the precision remained inconclusive, these biomarkers may serve as an alternative in limited-resource settings because, as cytokine/chemokine secreting cells, the changes in the ML ratio may contribute to the variation of cytokine/chemokine composition. Further consideration related to other progressive infections possibly affecting the results should be assessed in the implementation of the markers.

Despite substantial differences in immune responses between adults and pediatric patients, some similarities can be observed. There is an increase of lymphocyte and a decrease of macrophage percentages in both adults and children with TB (Herrera et al., 2022). Nonetheless, the downregulation of IL1RN and IL1R2, which are involved in the generation of the adaptive immune response, is believed to cause disseminated infections in pediatric patients (Bah et al., 2018). Even so, the common immune pathways activated during TB infection are similar between TB adults and children regardless of the different clinical symptoms presented by these two groups (Berry et al., 2010). Several immune chemokines have also been reported to differentiate between active TB and non-TB in children, which showed higher baseline levels of CCL1, CXCL1, and CXCL10 (Kumar et al., 2021a). Those chemokines can decrease with treatment and are useful for treatment monitoring.

In adult TB patients, chemokines have been widely considered as potential diagnostic biomarkers. Indeed, CCL1, CCL3, CXCL1, CXCL10, and CXCL11 were associated with TB disease severity and positively correlated with bacteria load, while CCL1, CXCL9, CXCL10, and CXCL11 could discriminate PTB from LTBI infection in individuals (Kumar et al., 2019b). MCP-1 was also demonstrated to differentiate between PTB, endobronchial TB, and TB pleurisy (Xiong et al., 2016). Baseline levels of plasma CCL2, CCL3, CCL4, CXCL8, CXCL10, CX3CL1, and CXCL1 could predict unfavorable treatment outcomes in PTB (Kumar et al., 2021b). Nevertheless, pregnant women have different immune profiles compared to adults, and transient changes in immunity occur during pregnancy (Abu-Raya et al., 2020). A previous study found that pregnant women with LTBI had lower levels of various proinflammatory cytokines, such as IL-1β, IL-6, and IL-17A, and a higher level of IFN-γ, notably during the second trimester (Naik et al., 2021). These results suggested that latent TB during pregnancy is characterized by a distinct immune profile with high levels of IFN-γ but low levels of other immune markers known to play a role in TB. The mechanisms for this finding were unclear. Previous studies on the mechanisms of TB disease showed that IFN-γ could give negative feedback to IL-1β, IL-6, and IL-17A in some instances (Nandi and Behar, 2011; Eigenbrod et al., 2013), thus high levels of IFN-γ might be associated with a low level of other markers. Additionally, an increase in neutrophil levels was observed during pregnancy and associated with lower IL-6 and IL-17 in TB infection (Sacks et al., 1998; Nishida et al., 2007).

Furthermore, immune markers might be altered by several factors, including age. Studies of immune marker concentration for diagnosis or monitoring TB disease in either elderly or preterm neonates were hard to find. However, physiological changes and the complicated clinical conditions that commonly appear in this population should be noted (Ciabattini et al., 2018; Aulin et al., 2021). TB in the elderly is often associated with the reactivation of quiescent lesions (Caraux-Paz et al., 2021). These activated lesions might be attributable to changes in the immune system related to aging, notably the decline in the ability to reactivate acquired immunity or immune response expressions. Elevated plasma levels of IL-6, IL-1, and TNF-α have been observed in the elderly (Weiskopf et al., 2009; Martín et al., 2017). These markers were previously described as “inflammaging” and are considerable for biomarker-based therapeutic optimization strategies (Franceschi et al., 2018; Aulin et al., 2021). On the other hand, preterm infants have a notable reduction in IL-1β, IL-6, and TNF-α because their innate immune responses are amid the developing process (Delanghe and Speeckaert, 2015). In addition, within several days after birth, preterm and term infants were reported to have higher levels of CRP and PCT, likely related to birth stress (Sharma et al., 2012).

Since TB frequently co-exists in immune-compromised patients, people who live with HIV are the most vulnerable to infection and disease progression risk (Organization, 2015). The increased susceptibility to TB in HIV-infected individuals is likely due to the characteristic immunological effects of HIV that predominantly infect CD4⁺ T cells (Okoye and Picker, 2013; Geremew et al., 2020). Both CD4⁺ and CD8⁺ T cells have been implicated in the immune response against Mtb infection (Prezzemolo et al., 2014). However, animal studies have shown that CD4⁺ T cells are more critical than CD8⁺ T cells in the host’s immune response to fight TB (Lin and Flynn, 2015). Additionally, there is growing evidence that TNF-α-dependent macrophage apoptosis is reduced with HIV-TB co-infection (Patel et al., 2007; Walker et al., 2013). These findings indicate that not only is adaptive immune response altered in HIV subjects, but also the innate immune response. A further condition regarding HIV patients that may complicate TB treatment is tuberculosis-immune reconstitution inflammatory syndrome (TB-IRIS). TB-IRIS is an abnormal and excessive immune response against alive or dead Mtb that usually occurs in HIV-infected patients (Tadokera et al., 2011; Naidoo et al., 2012). This condition may appear after the initiation of antiretroviral therapy independently of an effective suppression of HIV viremia. TB-IRIS decreases the levels of several circulating inflammatory markers, including IL-6 and Th1/2/17 cytokines (IL-2, IL-3, IL-12, IL-15, and IL-17A) (Ravimohan et al., 2015). The cytokines stimulate T and natural killer (NK) cell responses to activate macrophages and control Mtb infection (Gonzalez‐Juarrero et al., 2005; Rausch et al., 2006; Méndez-Samperio, 2010). Therefore, low pre-ART levels of these cytokines in TB-IRIS patients may implicate abnormal innate immune responses and further hamper pathogen clearance. Measurement of immunologic profiles prior to and early after ART initiation for HIV patients is encouraged to predict the treatment response and guide the regimen changes in TB therapy.

Another concern regarding immunocompromised patients is diabetes mellitus (DM). It has been widely reported that DM and TB often appear together, especially in the aging population (Niazi and Kalra, 2012; Lin et al., 2015). DM is known as a risk factor for the conversion of LTBI into active TB (Hensel et al., 2016). The function of neutrophils, macrophages, dendritic cells (DC), NK cells, and other components of innate immunity, as the first-encounter immune responses, are compromised due to metabolic changes in DM patients (Martinez and Kornfeld, 2014; Yew et al., 2017). Hence, type 2 DM patients may be characterized by the decreased secretion of IL-1β, IL-12, IL-18, and IFN-γ, which increases their susceptibility (Stalenhoef et al., 2008). Additionally, DM patients also manifest impaired adaptive immune responses. Previous studies revealed that TB patients with DM had significantly lower frequencies of both myeloid DC and plasmacytoid DC compared with individuals with only TB (Ayelign et al., 2019). Hyperglycemia was reported to cause a lower DC and neutrophils count in TB patients with DM (Kumar Nathella and Babu, 2017; Raposo-Garcia et al., 2017). As an explanation, hyperglycemia may increase adhesion and integrin expression, reduce chemotaxis, defect phagocytes, and reduce microbicidal activity (Raposo-Garcia et al., 2017; Ayelign et al., 2019).

Furthermore, DM also may potentially influence and/or decrease the frequencies of Th1 and Th17 cells due to the increased frequencies of Th2 cells which produce IL-4 (Yamashiro et al., 2005; Kumar Nathella and Babu, 2017). Given all the aforementioned conditions, the incorporation of immunomics into clinical settings would bring more insights to guide safe and effective treatment for these special populations. It is also worth noting that Mtb-specific immune responses are probably not homogenous in all populations and might be influenced by the immunity alterations of each specific population. Hence, using generalized potential immune biomarkers to monitor and diagnose TB in these populations may not be appropriate.

To overcome IIV, TDM has been utilized to adjust the dose following patient’s PK/PD profile. This can reach well-defined PK/PD targets associated with the efficacy and toxicity of drugs (Alsultan and Peloquin, 2014; Lange et al., 2020). Most of the dose adjustments that have been done in clinical settings follow the PK/PD indices of specific anti-TB drugs (Alsultan and Peloquin, 2014; Sturkenboom et al., 2021). Nonetheless, substantially unpredictable variations in treatment response remain for some patients. Demographic characteristics, clinical conditions, and genetic variants of metabolic enzymes or transporters of anti-TB drugs are commonly identified as the source of IIV affecting the treatment response of patients. Apart from the previous findings, other sources of IIV, in relation to pathogen characteristics, a patient’s immune response, or the effects of inflammation on PK, should also be explored (Young et al., 2020; Aulin et al., 2021). In the near future, precision medicine for TB should not only depend on the pharmacological aspects of anti-TB drugs, but rather also focus on biomarker-based treatment decisions and monitoring (Alsultan and Peloquin, 2014; Aulin et al., 2021). Biomarkers hold the potential to inform treatment strategies and aid decisions throughout all phases of infection (Delhalle et al., 2018). However, the current application of biomarkers in TB treatment has primarily been introduced for diagnosis.

Recently, immune profile biomarkers have been of interest to quantify treatment efficacy in relation to treatment response due to the length of time needed for bacterial culture and the difficulty of sputum production (Thorsted et al., 2020; Mensah et al., 2021). During TB infection, the immune response is triggered by Mtb, producing several immune biomarkers. To guide and optimize TB treatment, immune biomarkers should ideally reflect the underlying pathogen load. In addition to the role of immune biomarkers for treatment monitoring and outcome prediction, the effect of immune biomarkers on IIV of anti-TB drugs PK also should be evaluated. It is widely known that infection induces inflammation effects on the absorption, distribution, metabolism, and elimination processes of drugs (Radke et al., 2017; Charlton and Thompson, 2019). Hence, immune biomarkers that capture infection-induced inflammation effects and predict IIV on anti-TB drug PK parameters are significantly relevant for further individualizing TB treatments.

Early and accurate detection of active TB is crucial to prevent transmission of TB, yet existing diagnostic procedures are insufficient. Despite the development of several immunoassays for determining antibodies or antigens in serum samples, only a few diagnostic methods are available (Kashyap et al., 2007). Typically, ELISA is the primary analytical tool to quantify mycobacterium-specific components (Kashyap et al., 2007; Reither et al., 2009), host antibody response (Ben Selma et al., 2010; Zhu et al., 2012), and cytokine/chemokine production (Hur et al., 2013; Indrati et al., 2022) for diagnosis and monitoring treatment (Kim et al., 2021). Kashyap et al. employed an indirect ELISA method to diagnose TB by monoclonal antibodies against Ag 85 complex, which yielded high sensitivity and specificity of up to 82% and 86%, respectively. Recently, a quantitative lateral flow assay was established for adjunct diagnostics and treatment monitoring of TB and COVID-19 (Pierneef et al., 2022). Analysis of seven host sera proteins exhibited potential in differentiation of LTBI and TB, HC and COVID-19 patients, and severe COVID-19 and TB (Pierneef et al., 2022). Additionally, there are several alternative approaches to detect TB in serum and urine, including dot-immunobinding assay (Bentley-Hibbert et al., 1999), nucleic acid amplification (D’Amato et al., 1995), the application of two-dimensional polyacrylamide gel electrophoresis (2DGE), and liquid chromatography-tandem mass spectrometry (LC-MS) (Kashyap et al., 2005).

Flow cytometry is an alternative approach frequently used to evaluate various cell surface markers and intracellular cytokine production, permitting the characterization of distinct immune cell types. Several analyses using this method manifest the possibility of distinguishing between active, LTBI, and uninfected patients (Harari et al., 2011; Pollock et al., 2013; Estévez et al., 2020a), and assessing treatment efficacy (Ahmed et al., 2018) as well as drug susceptibility (Hendon-Dunn et al., 2016).

More recently, multiplex bead array assays have emerged in analytes quantification (Elshal and McCoy, 2006) simultaneously due to their capability to provide a vast quantity of bio-information from a small sample size for a better prognosis, diagnosis, and treatment (Ahsan, 2021). The multiplex immunoassays make use of conventional immunoassay methods that capture circulating antibodies or proteins by using either antibodies or peptides/proteins as binder molecules (Ahsan, 2021). In suspension experiments, the targeted ligands are immobilized on color- or size-coded plastic microspheres, and flow cytometry is used to identify particular fluorescent signals (Ellington et al., 2010). Bead-based flow cytometry technologies that enable multianalyte measurements in clinical diagnostics are already widely accessible on commercial platforms (Spindel and Sapsford, 2014). By using multiplex methods, such as xMAP technology developed by Luminex or Bioplex, immunomics is by far strengthened in reducing technical burden. The Luminex xMAP system is one of the pioneers (Luminex Corp., USA). It has been employed in numerous recent studies to investigate and validate immune biomarkers for TB diagnosis (Phalane et al., 2013; Xiong et al., 2016; Kumar et al., 2021a; Mutavhatsindi et al., 2021), treatment monitoring (Kumar et al., 2020), and recurrence prediction (Sivro et al., 2017). Conversely, a conventional ELISA experiment is laborious and usually takes days to quantify multiple immune molecules. However, a multiplex system facilitates simultaneous evaluation and demonstrates correlation in data with ELISA within several hours (Wang et al., 2005). On another scale, Meso Scale Discovery (MSD), a global multiplex analysis to characterize innovative assays for biological molecules, is also worth considering. Sharing comparable specifications, both Luminex and MSD are applicable for multiple immune profile analyses (Chowdhury et al., 2009). However, while MSD has a wider dynamic range and sensitivity, Luminex’s strength is emphasized in its precision (Sourial et al., 2009; Günther et al., 2020), which may be more suitable for the considerable input of biomarkers across the wide range of personalized treatments. Therefore, despite the need for further validation, multiplex technology, especially Luminex, is taking steps to unlock complex cytokine networks and reveal new biosignatures for TB. The remarkable milestones of these TB-related immunological investigations are briefly depicted in Figure 2 .