The diagnosis of latent tuberculosis infection (LTBI): currently available tests, future developments, and perspectives to eliminate tuberculosis (TB).

INTRODUCTION: Despite great efforts, tuberculosis (TB) is still a major public health threat worldwide. For decades, TB control programs have focused almost exclusively on infectious TB active cases.  However, it is evident that this strategy alone cannot achieve TB elimination. To achieve this objective a comprehensive strategy directed toward integrated latent tuberculosis infection (LTBI) management is needed. Recently it has been recognized that LTBI is not a stable condition but rather a spectrum of infections (e.g., intermittent, transient or progressive) which may lead to incipient, then subclinical, and finally active TB disease. AIM: Provide an overview of current available LTBI diagnostic test including updates, future developments and perspectives.
RESULTS: There is currently no test for the direct identification of live MT infection in humans. The diagnosis of LTBI is indirect and relies on the detection of an immune response against MT antigens, assuming that the immune response has developed after a contact with the biological agent. Tuberculin skin test (TST) and interferon gamma release assays (IGRAs) are the main diagnostic tools for LTBI, however, both present strengths and limitations. The most ancient diagnostic test (TST) can be associated with several technical errors, has limited positive predictive value, is being influenced by BCG vaccination and several conditions can reduce the skin reactivity. Notwithstanding these limitations, prompt identification of TST conversion, should orientate indications for preventive therapy of LTBI. IGRAs have superior specificity, are not affected by M. bovis, BCG vaccination and other environmental mycobacteria. However, they present some logistical and organisational constraints and are more expensive. Currently, the WHO guidelines recommend that either a TST or an IGRA can be used to detect LTBI in high-income and upper middle-income countries with estimated TB incidences less than 100 per 100,000 population. Two skin tests (C-TB and Diaskintest), using only two specific M. tuberculosis antigens (ESAT-6 and CFP-10) instead of the tuberculin solution, have recently been developed but, to date, none of these tests is available on the European market.
CONCLUSION: Early identification and treatment of individuals with LTBI is an important priority for TB control in specific groups at risk within the population: this is of crucial meaning in recently infected cases both at the community level and in some occupational settings. Currently there is no gold standard test for LTBI: an improved understanding of the available tests is needed to develop better tools for diagnosing LTBI and predicting progression to clinical active disease.

Introduction

Despite great efforts, tuberculosis (TB) is still a major public health threat worldwide, with 10 million new cases and 1.2 million deaths in 2018 (56). One fourth of the global population (approximately 2 billion persons) is estimated to be infected with Mycobacterium tuberculosis (MT) (11, 19), including approximately 13 million people in the United States (35).

Most infected individuals are asymptomatic and classified as having latent tuberculosis infection (LTBI). LTBI is defined by the Word Health Organization (WHO) as a state of persistent immune response to MT antigens, with no evidence of clinically manifest active TB (55); thus, this condition identifies the individuals who have being in contact with MT and have developed an immune response, but this does not imply necessarily the persistence of living pathogens in the human body (30). The vast majority of infected people have no signs or symptoms of TB and will never develop the disease.

The risk of developing active TB following infection depends on age, the quality of the immune defense mechanisms, and the time elapsed since the infection. The estimated liftetime risk is 5-10%, but it is higher in small children, immunocompromised individuals, and shortly (within 1-2 years) after a contact with a contagious TB case. LTBI is not a stable state; intermittent, transient or progressive episodes of mycobacterial replication can lead to incipient, then subclinical, and finally active TB disease (Fig. 1) (13).

Figure 1

The natural history of tuberculosis disease: pathways of progression; adapted from Drain et al. (13)

In high-income countries, the incidence of active TB disease has continued to decline over the recent decades, but the prevalence of LTBI has remained stable.

For decades, TB control programs have focused almost exclusively on infectious TB cases. Active case finding and proper management of active TB should be the top priority for control programs. However, it is evident that this strategy alone cannot achieve TB elimination. This important objective can be achieved through comprehensive strategies aimed at a proper and integrated LTBI and active TB management (27, 32). In fact, because a majority of new TB cases are a result of reactivation of remote LTBI rather than recent infection, intensification of LTBI screening and treatment strategies is recognized as a crucial component of TB elimination, especially in low TB prevalence settings. LTBI is a large reservoir for active TB and progression from untreated LTBI accounts for approximately 80% of U.S. TB disease cases.

The WHO action plan for TB elimination in low-incidence countries (28) (defined as those with a TB notification rate of <100 TB cases - all forms - per million population) highlighted the public health role of screening for active TB and LTBI in contacts of infectious cases. One of the priority actions in low TB incidence countries is the identification of contagious cases to implement contact tracing and prevent new incident cases of active TB.

Early diagnosis of LTBI after a documented or suspected exposure is key in order to assess the risk of active TB and to provide prompt preventive treatment. Therefore, detecting groups at higher risk of developing the disease is needed both at the community and occupational levels in low incidence countries in order to eliminate TB (17).

It has been reported that MTB transmission is more likely in some confined environments such as healthcare settings (42, 51), long-term care facilities, high-congregate settings (e.g., homeless shelters, prisons, schools) with an increased risk for patients, residents and workers alike. Indeed, this is valid in prison settings for both LTBI and TB (7, 9, 16, 26, 36), as well as in other congregate settings with asylum seekers and migrants from high TB incidence countries (5, 8, 12, 15, 21, 25, 29, 45). Poor hygiene conditions, poor ventilation, and high density of contagious residents and co-workers could explain the increased risk of MT transmission in these settings (37).

Occupational physicians and specialists in preventive medicine can play a key-role in the implementation of LTBI surveillance, early diagnosis, and preventive therapy in categories and individuals with a higher risk than the general population of being exposed to MTB and/or to develop active TB. In this view, TB risk-assessment in the occupational context needs to focus on the following items: (I) local epidemiology and working environment conditions (e.g., healthcare and specific high-congregate settings); (II) high risk-medical procedures (e.g., bronchoscopy, respiratory rehabilitation); and (III) individual risk factors associated with progression from LTBI to active disease (e.g., co-morbidities, iatrogenic immune depression).

This paper briefly summarizes the recent acquired knowledge on LTBI and current available diagnostic tools, also reporting future developments and challenges in the diagnosis. A better knowledge and comprehension of the available diagnostic tests for LTBI diagnosis, including their strengths and limitations, as well as their intergroup applicability, can be useful to properly identify and manage this condition in order to eliminate TB.

Available tests for LTBI

There is currently no test for the direct identification of living MT infection in humans.

The diagnosis of LTBI is indirect and relies on the detection of an immune response against MT antigens, asssuming that the immune response has developed after a contact with MT. Tuberculin skin test (TST) and interferon gamma release assays (IGRAs) are the main diagnostic tests for LTBI; both are indirect tests based on immune response to TB and do not directly assess the presence or viability of TB bacilli.

Furthermore, no test is currently available which would allow the distinction between an immune reaction due to LTBI from that induced by active TB.

Thus, the distinction between LTBI and active TB relies on clinical, bacteriological, and radiological findings.

Current available LTBI diagnostic test are reported and discussed below.

The tuberculin skin test (TST)

The classical test for the detection of LTBI is the TST, the most ancient diagnostic test still in use in medicine, which relies on the reaction against MT antigens injected in the dermis of the skin. The intensity of the local inflammatory reaction due to the release of cytokines by sensitized lymphocytes is measured after 48 to 72 hours and has been shown by some Authors to be correlated with the risk of future TB (20, 38).

TST is cheap, can be performed anywhere by any healthcare workers with adequate trainig without the need of technical equipment, except a cold chain for keeping the tuberculin solution. Nevertheless, the test is characterized by a large number of drawbacks which makes its use far from ideal. Moreover, tuberculin solution contains a large number of antigens (~200), most of which are common to M. tuberculosis, M. bovis, BCG, and environmental mycobacteria: a positive test reaction can, therefore, be elicited by a prior vaccination with BCG or a contact with environmental mycobacteria and has a low specificity (49). Furthermore, the technique for placing tuberculin and reading the reaction can be associated with several technical errors (e.g., injection in the wrong skin layer, fluid leaks, subjective evaluation of the skin reaction with intra- and inter-observer errors). Several conditions can decrease the skin reactivity, such as viral infections, immune depression, young and old age (Tab.1). Repeating the test can elicit an artificial increase in size of the second reaction (the so-called “booster effect”) (34). Furthermore, the test is performed without any positive or negative control and the quality of the tuberculin can also influence the proportion of reactions interpreted as positive (39).

Another limitation is the limited predictive value for TB disease. In other words, a majority of subjects with positive TST results do not progress to active TB disease, so overtreatment is inevitable, since there is no way to know which individual with a positive TST result will actually benefit from LTBI therapy. Notwithstanding these limitations, prompt identification of TST conversion, shortly after a contact with an infectious case, should orientate the indication for preventive therapy.

Interferon Gamma Release Assays (IGRAs)

An alternative test is the blood-based test relying on the in vitro measurement of gamma-interferon release by sensitized lymphocytes after stimulation with antigens from MT. Basically, the test uses the same principle as the TST (evaluation of the release of cytokines by sensitized lymphocytes). It needs a blood sampling, transportation and a laboratory equipment; it is more expensive than the TST but offers several advantages. The first and most important advantage is the fact that only 2 antigens from MTB (ESAT-6 and CFP-10), which are absent from M. bovis, BCG and most environmental mycobacteria, are used for stimulating the lymphocytes. Then, the test has a higher specificity. The second positive feature of the test is the standardization and objectivity of the procedure, without the subjective bias due to the injection technique and the reading of the skin reaction. Furthermore, the test is performed with a negative and positive control, thus decreasing the risk of error. The test can be repeated without the risk of boosting the result. Nevertheless, apart from cost and technical requirements, the performance of the test requires attention to several conditions, among which the transport conditions of the blood samples to the laboratory, the delay in processing, and the availability of a specialized laboratory (40).

Currently, two tests are available on the market (the QuantiFERON-TB Plus and the T-SPOT.TB test), which differ in the laboratory procedure but rely on the same principle.

QuantiFERON-TB Plus (QFT-Plus) is the new version of the QFT-Gold In-Tube. Compared to the QFT-Gold In-Tube, with antigens optimized to stimulate CD4+ T cells, QFT-Plus contains new antigens optimized for both CD4+ and CD8+ T cell stimulation (24). QFT-Plus has potential to indicate recent infection and disease activity. A recent meta-analysis has revealed that QFT-Plus is a more sensitive test compared to QFT-GIT for detecting M. tuberculosis infection (46).

The T-SPOT.TB is an enzyme-linked immunospot assay performed on separated and counted peripheral blood mononuclear cells, which include circulating monocytes and lymphocytes; it uses ESAT-6 and CFP-10 peptides. The result is reported as number of interferon-gamma-producing T cells (spot-forming cells in antigen wells minus negative control wells). An individual is considered positive for MT infection if the spot counts in the TB antigen wells exceed a specific threshold relative to the control wells (33). In general, IGRA results are reported as one of the following: positive, negative, indeterminate or uninterpretable results (some test manufacturers and laboratories add a borderline category for interpretable test results close to the cut-off).

Skin tests with specific Mycobacterium tuberculosis antigens: C-TB and Diaskintest

Two skin tests (C-TB and Diaskintest) using only two specific MT antigens (ESAT-6 and CFP-10) instead of the tuberculin solution have recently been developed. They have a specificity similar to the IGRAs (no influence of prior BCG vaccination or environmental mycobacteria) but must be placed and read with a technique similar to the TST (intradermal injection, reading after 48-72 hours).

These tests seem to have performance similar to the IGRAs (2, 44, 47). None of the tests is currently available on the European market.

Comparison between available diagnostics for LTBI

Several studies have compared the performance of the TST and the IGRAs in adults and children (3, 18, 31). Globally, these studies confirm that the concordance is satisfactory in patients with active TB or patients with recent TB contact, but a large group of adults and children has discordant test results. The combination of positive TST and negative IGRA is frequently associated with prior vaccination with BCG or exposure to environmental mycobacteria (false-positive TST), whereas the combination of negative TST and positive IGRA is less frequent and may be attributed to a decrease in the reactivity of the immune system associated with very young or very old age or immmuno-depression (false-negative TST). The superior specificity of IGRAs compared with TST is demonstrated by the very low risk of TB development in subjects with a negative IGRA, independently of the TST result (4). The performance of the IGRA is satisfactory in children from the age of 2 years (10). Systematic performance of the IGRA in subjects with a low-risk of infection (e.g., routine annual testing in healthcare workers) gives rise to variations in test results, with apparent conversions and reversions, mostly due to variations around the cut-off (6).

The intensity of the reaction to TST and IGRA seems to be correlated with the risk of future TB (36, 54) but both tests have a low positive predictive value. The negative predictive value is high for both tests, but higher for the IGRA than for the TST (1).

A detailed comparison between the available diagnostic tests for LTBI infection is outlined in Table 2.

Table 2

A comparison of available diagnostics for latent tuberculosis (TB) infection, adapted from Pai M, Sotgiu G. (41)

Characteristic	*PPD-based tubercolin skin test	New specific skin test (under either development or validation)	IFN-γ release assays
Testing format	Intradermal skin test	Intradermal skin test (in vivo)	Ex vivo assay (ELISA or ELISPOT)
Antigens used	Purified protein derivate	ESAT-6 and CFP-10	ESAT-6 and CFP-10
Intended use	Screening for LTBI	Screening for LTBI	Screening for LTBI
Sensitivity	High	Modest	Modest
Sensitivity in immune-compromised population	Reduced	Reduced	Reduced
Specificity	Modest	High	High
Impact of BCG on specificity	High (when BCG is given after infancy or several times)	None	None
Ability to predict progression to active TB	Modest	Unknown (but likely modest based on indirect evidence from IGRAs)	Modest
Ability to resolve the various stages within the spectrum of M. Tuberculosis infection	Low	Low	Low

* Purified Proteine Derivative

With regard to LTBI diagnosis in high income and upper middle-income countries with estimated TB incidences less than 100 per 100,000 population, the current WHO Guidelines recommend that either a TST or an IGRA can be used to detect LTBI.

Future developments

The immunological sensitization induced by the contact with MTB is not limited to the release of Interferon-Gamma. Numerous other cytokines are released by several blood cells and can be measured in vitro (22).

An interesting study proved that the proportion of monocytes in blood and an elevated monocytes/lymphocytes ratio could be an indicator of the risk of development of TB after TB contact in patients with a positive TST (43).

Recently, several publications focused on the RNA gene signature or on the assessment of the genetic changes induced by mycobacteria. Such changes have also been observed after infection with other virus and bacteria (48). Changes seem to correlate with the infection and with the risk of developing TB, thus allowing to predict which infected persons will progress to TB (52, 57). Most tests use a combination of genetic markers to detect the infection and the risk of TB. The number of genetic changes assessed by diverse methods lies between 3 and 393 and the technology for their determination is still not available outside research laboratories. Currently, a 3-gene signature has been demonstrated as promising (53). A recent study assessed the diagnostic accuracy of several RNA sequencing tests in patients with symptoms compatible with pulmonary TB (50). Four of them met the minimum WHO requirements for a triage test but not for a confirmatory test.

Apart from blood tests, other tests may allow for the detection of the presence of living mycobacteria in the body before the emergence of active disease. A fascinating study by Esmail demonstrated that 2-deoxy-2-(18F) fluoro-d-glucose PET-CT (FDG-PET-CT) can detect living mycobacteria in the lung and lymph nodes of HIV-positive patients and predict the progression to TB (14).

Conclusion

LTBI is a complex and heterogeneous state resulting from the dynamic interaction between MT and the host’s immune response. Early identification and treatment of individuals with LTBI are priorities for TB control in specific groups at risk within the population: this is of crucial meaning in recently infected individuals both at the community level and in some occupational settings.

Screening for LTBI and proper preventive treatment are key-elements of the End TB strategy (53).

The currently available diagnostic methods for the diagnosis of LTBI are the “century-old” TST and, since 2005, the IGRA test. The IGRAs provide an opportunity for more targeted LTBI treatment than offered with TST, but the implementation of costly and laboratory-intensive IGRA testing remains limited or not available in many settings.

As there is no gold standard test, an improved understanding is needed in this field, also in order to develop better tools for diagnosing LTBI, distinguishing it from active TB, and predicting progression from LTBI to clinical disease (23).

Table 1

List of possible reasons accounting for tuberculin skin test (TST) misleading results

FALSE POSITIVE REACTIONS	FALSE NEGATIVE REACTIONS
Infection with non-tuberculosis mycobacteria	Cutaneous anergy (inability to react to skin test because of weakened immune system)
Previous BCG vaccination	Recent TB infection (within 8-10 weeks of exposure)
Incorrect method administration	Ancient TB infection (many years)
Incorrect interpretation of reaction	Very young or very old age
Incorrect bottle of antigen used	Recent live-virus vaccination (e.g., measles and smallpox)
	Overwhelming TB disease
	Some viral illnesses (e.g., measles and chicken pox)
	Incorrect method of administration
	Incorrect interpretation of reaction (intra- and inter-observer errors)

Tabella 1

Elenco dei possibili motivi di una non corretta interpretazione dei risultati del test cutaneo alla tubercolina (TST)

Reazioni false positive	Reazioni false negative
Infezione da micobatteri non tubercolari	Anergia cutanea (incapacità di reazione al test cutaneo a caua d’immunocomprosissione)
Pregressa vaccinazione con BCG	Infezione da TB recente (entro 8-10 settimane dall’esposizione)
Errata tecnica di somministrazione	Infezione da TB pregressa (molti anni)
Interpretazione errata della reazione	Fasce estreme di età (molto giovane e avanzata)
Utilizzo di flacone di antigene errato	Recente vaccinazione con virus vivente attenuato (es., morbillo e vaiolo)
	Malattia tubercolare diffusa
	Alcune malattie virali (es., morbillo e varicella)
	Errata tecnica di somministrazione
	Interpretazione errata della reazione (errori intra e inter-osservatore)

Tabella 2

Comparazione tra i test diagnostici disponibili per Infezione tubercolare latente, adattata da Pai M, Sotgiu G.

Caratteristica	TST	Nuovi test cutanei specifici	IGRAs
Formato del test	Test cutaneo intradermico	Test cutaneo intradermico (in vivo)	Ex vivo assay (ELISA o ELISPOT)
Antigeni utilizzati	Derivato proteico purificato	ESAT-6 e CFP-10	ESAT-6 e CFP-10
Destinazione d’uso	Screening per ITBL	Screening per ITBL	Screening per ITBL
Sensibilità	Alta	Modesta	Modesta
Sensibilità nella popolazione immuno-compromessa	Ridotta	Ridotta	Ridotta
Specificità	Modesta	Alta	Alta
Impatto del BCG sulla specificità	Alta (quando il BCG viene somministrato dopo l’infanzia o più volte)	Nessuno	Nessuno
Capacità di prevedere la progressione verso TB attiva	Modesta	Sconosciuta (ma probabilmente modesta in base a evidenza indiretta ottenuta con IGRAs)	Modesta
Capacità di definire le varie fasi nello spettro dell’infezione da MT	Bassa	Bassa	Bassa