These are in vitro blood tests that measure either the amount of IFN-γ released or the number of T cells releasing IFN-γ in response to specific antigens of M. tuberculosis complex (ESAT-6 and CFP-10). These antigens are absent from the Mycobacterium bovis BCG vaccine and most non-tuberculous mycobacteria, except for Mycobacteriummarinum, Mycobacterium kansasii, Mycobacterium szulgai and Mycobacterium flavescens. Several commercial kits are available [49] (Table 1), the most widely studied being:

• a)

  QuantiFERON-TB Gold Plus (QFT-Plus) [50–52]. This test is based on four tubes, two of which contain specific polypeptide antigens (CFP-10 and ESAT-6) that stimulate CD4+ and CD8+ T cells, potentially relevant for identifying recent infections or active infectious states. The response is measured using ELISA or chemiluminescence immunoassays [53].

• b)

  T-SPOT.TB [54]. This test quantifies IFN-γ-producing T cells, and readings are based on ELISPOT assays.

Results are interpreted as positive, negative or indeterminate, based on the response to stimulation with specific antigens, taking into account the internal positive and negative controls.

Indeterminate results are uncommon and can be due to: (1) pre-analytical or analytical errors, or failure of the positive and/or negative controls, and (2) patient-related factors: viral infections that impair cellular immunity (e.g. HIV with low CD4 T cell counts), immunosuppressive medications (corticosteroids, chemotherapy, biologics), autoimmune and/or haematological malignant disease, chronic conditions (diabetes, chronic kidney failure, advanced liver disease) or extreme age.

The main strength of IGRAs is their high specificity, as results are not affected by prior BCG vaccination, and only a few non-tuberculous mycobacterial species interfere with the results. Moreover, IGRAs have been shown to be more sensitive than TSTs in immunosuppression, slightly more so with the ELISPOT assay [55], they require only a single patient visit, and provide objective results.

Their main limitations relate to their higher cost, the need for blood sampling and the requirements for laboratory facilities and specific technical training. In young children, their use is limited due to the weaker supporting evidence and higher rates of indeterminate results. As with TSTs, the sensitivity may be lower in immunosuppressed individuals, and the results cannot distinguish between TBI and TB [56–58].

The TST is the oldest diagnostic test currently recommended. It measures the delayed-type hypersensitivity response mediated by T cells, through an intradermal injection of a purified protein derivative (PPD) of M. tuberculosis into the upper third of the flexor surface of the forearm (Mantoux method) [59].

For interpretation, the diameter of the induration (skin thickening) should be measured rather than the erythema. Although different cut-off points have been accepted depending on individual and epidemiological risk factors for developing the disease [60], the test should be performed in people for whom a positive result would prompt preventive treatment. In this context, an induration of ≥5mm would be the main outcome used as a cut-off for recommending TPT. Higher cut-offs (≥10 or ≥15mm) may be appropriate in epidemiological studies or in individual cases where the efficacy of TPT has not been sufficiently demonstrated.

The main strengths of the TST are its low cost, wide availability, and performance at the point of care, without the need for laboratory facilities. In addition, the test is supported by almost a century of evidence of its utility, particularly in children.

Its limitations are related to the lack of specificity, with positive reactions occurring due to prior BCG vaccination or exposure to non-tuberculous mycobacteria. Further, it has variable sensitivity, which is influenced by immunosuppression. The test has no internal positive and/or negative controls. Operationally, it requires two patient visits (one for the injection, and another 48–72h later, for reading the response), and the results are more subjective and less reproducible than those of IGRAs.

In recent years, skin tests have been developed and marketed that, unlike the TST, only contain highly specific antigens of the M. tuberculosis complex. Consequently, their specificity is higher than that of the TST [61–64]. They are also performed using the Mantoux method, with the transverse diameter of the induration measured after 48–72h. The results are interpreted according to the same criteria as the TST [65].

Various commercial tests are available including:

• a)

  Cy-Tb (SIILTIBCY®) formerly known as the C-Tb test. It contains two recombinant proteins, ESAT-6 and CFP-10. It is the only test currently approved and marketed in Spain.

• b)

  Diaskintest®. It is based on a recombinant protein produced using a genetically modified culture of Escherichia coli BL21 (DE3)/pCFP-ESAT.

• c)

  C-TST. The active ingredient is a recombinant fusion protein ESAT-6–CFP-10, expressed in genetically modified E. coli.

Alongside numerous improvements in established testing kits, rapid or ultra-rapid IGRAs are being developed, using miniaturised microfluidic systems and flow cytometry-based platforms to achieve same-day results (within 24h). In addition, researchers are exploring immunological biomarkers and molecular and transcriptomic tests, with higher positive predictive values for disease progression, but these are not yet available for clinical use [49,55,66–68]. The ultimate goal is to develop a test that not only diagnoses TBI but also identifies individuals at highest risk of progression to TB disease, thereby enabling more targeted and efficient TPT.

The diagnosis of TBI requires integrating microbiological, clinical and epidemiological factors. In high-prevalence regions, which are often those with fewer economic resources, WHO recommends prioritising TSTs as the initial screening tool, with subsequent confirmation by IGRA (for example, in individuals previously vaccinated with BCG) [65,69,70].

In low-prevalence regions, which typically have greater economic and social resources – such as Spain, the recommendation is to perform an IGRA, particularly in people who have received the BCG vaccine. In certain circumstances, a combination of TST and IGRA may be considered, or even the simultaneous use of two different IGRA techniques, to enhance diagnostic sensitivity [65,71–73].

Recent evidence indicates that a negative IGRA or TST result has an almost 100% negative predictive value for the development of disease. This requires that testing is technically well performed, before the initiation of any immunosuppressive or immunomodulatory treatment, and in the case of contacts, at least 8–12 weeks after exposure (window period) [56,70,74–78].

Several studies have also demonstrated the superiority of IGRAs over TSTs in predicting the development of the disease [69,74,79–81].

In children <5 years of age (and especially in those <2 years of age), TST (or TBST) plus IGRA is recommended to achieve maximum diagnostic sensitivity. In immunocompetent individuals>5 years of age, either TST (or TBST) or IGRA may be used. IGRAs or TBSTs, are particularly recommended for children who have received BCG vaccination [82].

Pending further data on the role of new antigen-specific skin tests, this guideline supports the use of IGRAs as the initial diagnostic test, reserving TSTs for settings with limited access to IGRAs or operational constraints, such as large-scale community contact tracing studies (e.g. schools, universities, or workplaces). In situations where higher diagnostic sensitivity is needed, both tests may be performed.

In any case, it should be noted that no TBI diagnostic technique, whether alone or in combination, can guarantee with absolute certainty the prevention of tuberculosis disease [83]

It is estimated that 5–15% of individuals infected with M. tuberculosis complex will develop TB disease during their lifetime. Therefore, when selecting populations to receive TPT, emphasis should be placed on those groups at the highest risk of progression to disease. This requires consideration of the context in which testing is performed and the strategy applied, whether through contact tracing or population screening.

In the context of a TB control programme, it is essential to consider the number of infected individuals who must be treated to prevent one case of disease, together with the efficacy of, and adherence to, preventive treatment. In short, the aim is to maximise the efficiency of this intervention. Accordingly, the highest-priority group for evaluation is contacts of patients with pulmonary TB, in whom the risk of progression to disease is markedly increased, particularly among household and close contacts. This risk varies as a function of age, being highest in children under 5 years, although in older individuals it may reach up to 28% [84–86]. Therefore, all contacts of a pulmonary TB case should be prioritised for TPT, irrespective of age, especially if they have shared living space for more than 3–6h a day.

In contacts at high risk of rapid progression to TB disease (children under 5 years or individuals with severe immunosuppression), initiation of TPT should be considered even in the presence of a negative IGRA or TST result. In individuals exposed to patients with pulmonary TB who initially test negative by TST or IGRA, testing should be repeated 8–12 weeks after the latest exposure; TPT is indicated if the result subsequently becomes positive [77,87,88]. Reverse contact tracing (to identify the source of infection) should be performed in cases of childhood TB, including extrapulmonary forms [27].

In clinical practice, the risk of TB development must be balanced against the potential toxicity of TPT. Table 2 summarises the risk factors for developing TB in infected individuals, classified as high, intermediate, or low risk [89]. In this context, screening for TBI should be performed in individuals at high-risk, prioritising people living with HIV, as well as the following groups: patients starting treatment with TNF inhibitors, those with end-stage renal disease, solid-organ or haematological transplant recipients, and individuals with silicosis, where the benefit of TPT clearly outweighs the associated risks [7,90,91]. It has also been suggested that the presence of fibrotic pulmonary lesions≥2cm2 in infected individuals with no previous TB treatment should be an indication for TPT [92–95]. TBI testing may also be considered on a case-by-case basis in individuals receiving biological agents such as IL-6R or JAK inhibitors, where the risk of developing TB disease is not negligible [96,97].

In other scenarios, such as patients with diabetes, gastrectomy, alcohol misuse, solid-organ or haematological malignancies, or long-term corticosteroid therapy, the benefits of TPT are less certain; therefore, TBI screening and treatment should be considered on an individual basis [98,99]. In individuals diagnosed with TBI during contact tracing of a patient with pulmonary TB who decline TPT, clinical and radiological follow-up should be undertaken every 6–12 months for two years (the period of highest risk of progression to TB disease after infection). [100,101]

In frontline healthcare workers, people deprived of liberty, migrants from high TB-prevalence countries (particularly minors and those who have arrived within the past five years), people experiencing homelessness and individuals with substance use disorders may also be considered for TPT on a case-by-case basis [7,38,102]. The decision to recommend TPT in patients previously treated for TB or TBI should likewise be individualised. In such situations, TBI tests are usually positive and cannot distinguish whether reinfection has occurred. It should also be noted that individuals with a prior episode of TB are at increased risk of developing disease again if reinfected with M. tuberculosis[103,104].

Before starting preventive treatment, it is essential to rule out TB disease.

In immunocompetent individuals without symptoms suggestive of TB (cough, haemoptysis, fever/low-grade fever, weight loss, night sweats), chest radiography has a sensitivity exceeding 90–95%; therefore, the absence of lesions can be reasonably used to rule out pulmonary TB [105].

In people living with HIV, symptoms have a sensitivity of 83%, but low specificity (38%), which is not improved by radiology (20%). However, when combined with C-reactive protein levels>5mg/L, specificity increases to 76%, and to 90% in outpatients on antiretroviral therapy [105,106].

Fig. 1 shows an algorithm for ruling out TB disease in infected individuals. It highlights the performance of a chest X-ray to exclude TB. Rapid molecular techniques (real-time PCR and others) should be considered in individuals with symptoms and/or radiological findings suggestive of TB [27].

Fig. 1.

Algorithm to rule out tuberculosis in asymptomatic individuals. * Consider diagnostic evaluation in patients with severe immunosuppression, or if C-reactive protein>5mg/L in patients living with human immunodeficiency virus.

The first evidence suggesting that TPT could be effective was published more than 50 years ago, demonstrating the efficacy of long-term H monotherapy [92,107–117]. At that time, H was the only drug with both strong activity and good tolerability against the M. tuberculosis complex, and all studies conducted over the following three to four decades were based on H. Numerous clinical trials and publications consistently showed its efficacy when administered for six to nine months.

Although maximum efficacy was achieved with 9–10-month regimens provided adherence was good [118,119], field studies demonstrated that 6 months was similarly effective, as longer regimens of 9–12 months were associated with higher treatment discontinuation rates [92,118,119]. Based on this evidence and some subsequently published meta-analyses [120], the tendency for decades has been to recommend a 6-month regimen with H [121–125].

Nonetheless, it has been known for decades that the main activity of H lies in its bactericidal properties [126–128], acting mainly against continuously replicating bacilli, while it has limited sterilising effect against bacilli in a latent or semi-latent state. Given this, H cannot be considered the best drug for TPT, since its efficacy depends on prolonged administration over many months (to allow the bacilli in a latent/semi-latent state to replicate).

An ideal TPT drug should have sterilising activity, since this allows treatment durations to be shortened [126–128]. In recent years, numerous randomised clinical trials have evaluated such drugs, mainly rifamycins, particularly P and R [113,115,120,129–139]. These studies have demonstrated that the efficacy of 3-month daily R combined with H (3HR) [113,115,129–132], 4-month daily R [113,120,133,134], 3-month weekly HP [135–138] and 1-month daily HP [139,140] is similar to 6–9 months of H monotherapy, with the major advantage of shortening treatment duration, which improves adherence. However, these regimens may be associated with drug-drug interactions, as rifamycins are potent inducers of cytochrome P450.

Table 3 lists the TPT regimens that have been shown to be effective and safe, together with the recommended doses by age group, with some comments on their indication [7,113,115,120,129–139].

Based on its shorter duration, the 1HP regimen (daily dosing) would be the preferred option, particularly for people living with HIV. At present, however, it is not recommended for individuals under 14 years of age, as this age group has not been included in clinical trials, and therefore no scientific evidence is available, although it is reasonable to assume that it may also be effective [7,139,141,142]. The second option for TPT would be the 3HP regimen (weekly dosing), which requires only 12 doses in total, a feature that may facilitate adherence and supervised administration when deemed necessary. Although the WHO advises against the use of P during pregnancy due to the lack of evidence regarding its non-teratogenicity [7], some studies have reported a safety profile comparable to that of other TPT regimens [143], and its use in pregnant women may therefore be considered [144]. At the time of writing this report, P is not available in Spain or many other European countries, making the recommendation of 1HP or 3HP regimens unfeasible. Accordingly, the regimen of choice in Spain remains 3HR, with 6-9H reserved as an alternative in cases where R cannot be used due to toxicity, drug interactions, or mono- resistance to R. A study in Spain found the 3HR regimen to be the most cost-effective [145].

In individuals who have been in contact with TB patients with mono- or polyresistance to H, or intolerance to it, the treatment of choice for TPT is 4R [7,113,120,133,134]. In contacts of patients with multidrug-resistant TB (MDR-TB, resistant to at least H and R), 6 months of levofloxacin is recommended [7,146–148]. In this regard, there is no evidence for the efficacy of any TPT for contacts of patients with pre-extensively drug-resistant TB (MDR-TB that is also resistant to at least fluoroquinolones), and therefore these individuals should be managed on a case-by-case basis. Delamanid may be a good option given its mechanism of action [128].

The effectiveness of TPT in preventing TB disease is directly related to adherence. It is therefore essential to encourage each patient to fully adhere to the prescribed treatment throughout the diagnosis and treatment processes, especially among vulnerable populations [149]. Specialised TB units have achieved high rates of TPT adherence [150–153].