Mycobacterium tuberculosis-specific cytokine biomarkers to differentiate active TB and LTBI: A systematic review

Highlights

• Current diagnostic tests do not discriminate active TB from latent TB infection.

• Several novel cytokines show promise as stage-specific markers of TB infection.

• Inconsistent results across studies likely resulted from heterogeneous study design.

• Combinations of cytokines may perform better than any single cytokine alone.

• Further well-designed studies are needed for development of next generation tests.

Abstract

Objectives

New tests are needed to overcome the limitations of existing immunodiagnostic tests for tuberculosis (TB) infection, including their inability to differentiate between active TB and latent TB infection (LTBI). This review aimed to identify the most promising cytokine biomarkers for use as stage-specific markers of TB infection.

Methods

A systematic review was done using electronic databases to identify studies that have investigated Mycobacterium tuberculosis (MTB)-specific cytokine responses as diagnostic tools to differentiate between LTBI and active TB.

Results

The 56 studies included in this systematic review measured the MTB-specific responses of 100 cytokines, the most frequently studied of which were IFN-γ, IL-2, TNF-α, IP-10, IL-10 and IL-13. Ten studies assessed combinations of cytokines, most commonly IL-2 and IFN-γ. For most cytokines, findings were heterogenous between studies. The variation in results likely relates to differences in the study design and laboratory methods, as well as participant and environmental factors.

Conclusions

Although several cytokines show promise as stage-specific markers of TB infection, this review highlights the need for further well-designed studies, in both adult and paediatric populations, to establish which cytokine(s) will be of most use in a new generation of immunodiagnostic tests.

Introduction

It is estimated that one-quarter of the world's population is infected with Mycobacterium tuberculosis (MTB).¹ Despite effective treatment being available for over 70 years, it continues to be the leading cause of death from a single infectious disease worldwide². One of the main factors contributing to this toll is the limitations of current diagnostic tests. As a result, an estimated 3 million cases of tuberculosis (TB) each year are not diagnosed or notified.^{3, 4}

The diagnosis of active TB is made by visualisation of acid fast bacilli by smear microscopy and growth of MTB on culture, which is time consuming and has low sensitivity in children who have paucibacillary TB disease.^{5, 6} Nucleic acid amplification methods, such as GeneXpert MTB/RIF, are expensive and impractical in many resource-poor settings.^{7, 8}

Latent TB infection (LTBI) is diagnosed by either the tuberculin skin test (TST) or by an interferon (IFN)-γ release assay (IGRA), which are based on detecting an immune response to MTB antigens.^{9, 10} A TST result can be affected by prior BCG vaccination, potentially producing a false-positive result, as well as by other factors such as immunosuppression or malnutrition, which can produce a false-negative result.¹⁰ IGRAs measure a T-cell immune response, through IFN-γ production, to MTB-specific antigens (early secretory antigenic target (ESAT)−6, culture filtrate protein (CFP)−10 +/- TB7.7).⁹ They are not affected by prior BCG-vaccination, but have lower sensitivity in children, particularly those under five years of age.^{11, 12}

Neither the TST nor IGRAs can differentiate between active TB and LTBI.¹³ This distinction is crucial to determine appropriate treatment. A patient presenting with symptoms of active TB (e.g. cough, fever, weight loss, fatigue) may have a positive TST or IGRA result because they have active TB, or because they have LTBI and an intercurrent infection with a non-tuberculous respiratory illness. This is particularly a problem in countries where TB is endemic.

The World Health Organization (WHO) ‘End TB Strategy’ has called for the development of new diagnostic tests to address the shortcomings of existing tests to enable more accurate diagnosis of both LTBI and active TB.¹⁴ A growing number of studies have investigated the performance of novel MTB-specific cytokine responses, either alone or in combination, to improve the sensitivity and specificity of immunodiagnostic assays. These are envisaged as potential candidates for inclusion in novel immunodiagnostic tests for TB infection, and a number of studies have looked at their potential role as a stage-specific marker (i.e. whether a patient has active TB or LTBI).

This review summarises the results of all studies in humans that have investigated MTB-specific cytokine responses as diagnostic tools to differentiate between LTBI and active TB.

This review was done in accordance with the ‘preferred reporting items for systematic reviews and meta-analyses’ (PRISMA) statement.¹⁵ A systematic search of MEDLINE and EMBASE was done from 1946 to November 2019 using the OVID interface with the following search terms: (exp Mycobacterium tuberculosis/an, ch, im, ph OR exp Tuberculosis/bl, di, im, mi, ph, pp) AND (exp “intercellular signalling peptides and proteins” OR exp cytokines OR exp chemokines OR exp interferons OR exp interleukin 1 receptor antagonist protein OR exp interleukins OR exp lymphokines OR exp monokines OR exp tumour necrosis factors). Reference lists of identified relevant publications were also hand-searched.

Studies were included if they: 1) assessed the use of one or more cytokines (including chemokines) or soluble markers to discriminate between active TB and LTBI in humans, and 2) measured cytokines, chemokines or soluble markers in blood after stimulations with the MTB-specific peptides ESAT-6 and/or CFP-10. Studies were excluded if: 1) more than 10% of included patients were known to be immunosuppressed (e.g. HIV infection), as the immune response to TB infection may be altered in this group, and 2) diagnostic groups were not clearly defined, particularly where the control group included healthy household contacts (a group that may comprise TB-infected and TB-uninfected individuals). Meeting abstracts, studies without original data, studies not in English, and studies that evaluated IFN-γ alone were also excluded, as were studies that used flow cytometry due to the expensive and labour-intensive nature of this diagnostic modality.¹⁶

The search identified a total of 9670 studies. After removal of duplicates, 8512 studies were screened using titles and abstracts, of which 56 fulfilled the inclusion criteria16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 (Fig. 1). These 56 studies reported the results from 1848 patients with active TB and 1633 patients with LTBI. The results of these studies are summarised in Supplementary Tables 1A-C.