An Update on the Diagnosis of Tuberculosis Infection - ATS Journals

Pulmonary Perspective An Update on the Diagnosis of Tuberculosis Infection Luca Richeldi Respiratory Disease Clinic, Department of Oncology, Hematology, and Respiratory Disease, University of Modena and Reggio Emilia, Modena, Italy

Targeted testing and treatment of individuals with latent tuberculosis infection at increased risk of progression to active disease is a key element of tuberculosis control. This strategy is limited by the poor specificity of the tuberculin skin test in populations vaccinated with bacille Calmette-Gue´rin and its low sensitivity in immunosuppressed persons, who are at highest risk of progression. Two blood tests (T-SPOT.TB and QuantiFERON-TB Gold), based on detection of IFN-␥ released by T cells in response to M. tuberculosis–specific antigens, may offer an improvement on the skin test. However, validation is challenging due to the lack of a diagnostic gold standard. This critical appraisal of published evidence summarizes the diagnostic accuracy of the new tests. The blood tests have operational advantages over the skin test because no return visit is required, results are available by the next day, and repeated testing does not cause boosting. Both tests are significantly more specific than the skin test in populations vaccinated with bacille CalmetteGue´rin. The data suggest that T-SPOT.TB may be more sensitive than the skin test. Data in groups at high risk of progression to disease are scarce, and more research is needed in these populations, but it is clear that T-SPOT.TB performs better than the skin test in young children and HIV-infected people with active tuberculosis. Incorporation of these tests into programs for targeted testing of latent tuberculosis infection will reduce false-positive and falsenegative results inherent in tuberculin testing, equipping clinicians with more accurate tools for tuberculosis control and elimination in the 21st century. Keywords: blood test; diagnosis; interferon-␥; tuberculosis

LIMITATIONS OF THE TUBERCULIN SKIN TEST AND IMPACT ON STRATEGIES FOR TUBERCULOSIS CONTROL Targeted tuberculin testing for latent tuberculosis infection (LTBI) is a key component of tuberculosis (TB) control. It is based on identification and treatment of persons infected by Mycobacterium tuberculosis (MTB) who are at high risk for progression to active disease (1). This strategy is powerful because preventive treatment of latently infected people diminishes the risk of subsequent development of active TB by about 90% (2). The determinants of increased risk of progression to disease are recent infection with MTB and several host-related factors, all of which are associated with an impaired cell-mediated immune response. These include physiologic factors (e.g., young age, especially children aged under 5 years), pathologic factors (e.g., HIV coinfection and other conditions, including chronic

(Received in original form September 28, 2005; accepted in final form June 16, 2006 ) Correspondence and requests for reprints should be addressed to Luca Richeldi, M.D., Ph.D., Department of Oncology, Hematology, and Respiratory Disease, University of Modena and Reggio Emilia, Policlinico Hospital, Via del Pozzo, 71-41100 Modena, Italy. E-mail: [email protected] Am J Respir Crit Care Med Vol 174. pp 736–742, 2006 Originally Published in Press as DOI: 10.1164/rccm.200509-1516PP on June 23, 2006 Internet address: www.atsjournals.org

renal failure), and iatrogenic immunosuppression (e.g., anti– tumor necrosis factor-␣ agents, organ transplantation, and systemic corticosteroids) (3). People included in these vulnerable groups have more severe forms of TB that are often disseminated and fatal if untreated or treated late. The classic diagnostic tool for LTBI is the tuberculin skin test (TST), also known as the intradermal Mantoux test since 1910. It is the oldest diagnostic test in use in modern medical practice, and its limitations constitute the weakest element in the strategy of targeted testing of LTBI. The high-risk groups that are targeted for preventive therapy are also those in which the TST most often fails to detect LTBI (1, 4). Thus, the poor sensitivity of TST has a negative effect on the management of individuals who would benefit the most from targeted testing and preventive treatment. This same limitation of the skin test applies to its use as a diagnostic aid in the evaluation of cases of suspected active TB, when microbiological confirmation is awaited or not possible. Because infection is a necessary prerequisite for active disease, a sensitive test for MTB infection would help to rule out a diagnosis of active TB, particularly in immunosuppressed patients. However, due to its poor sensitivity, a negative TST in these patients is almost invariably clinically unhelpful and therefore not recommended by current guidelines (5). The second major drawback of TST is its low specificity. Because protein-purified derivative (PPD) is a culture filtrate of tubercle bacilli (6) containing over 200 antigens shared with the bacille Calmette-Gue´rin (BCG) vaccine and most nontuberculous mycobacteria (4), individuals vaccinated with BCG but not infected with MTB can test falsely positive using the tuberculin test. BCG is the most widely used vaccine in the world, and more than 3 billion people have received it (7). Even large TST reactions in adults living in low-prevalence areas can be due to a prior BCG vaccination (8), and a meta-analysis showed that BCG administration increases the likelihood of false-positive TST results for up to 15 years after vaccination (9). Nonetheless, some guidelines recommend ignoring the effect of BCG on TST (1, 10). Although setting a lower TST threshold tends to increase the sensitivity of tuberculin testing, it would do so at the cost of considering for preventive treatment a large number of people who are not infected but have positive TST reactions. As the prevalence of LTBI declines in low-prevalence countries, an increasing proportion of positive TST results will be due to prior BCG vaccination, making the low specificity of the tuberculin test a major obstacle to reach the goal of TB elimination.

THE BLOOD ASSAYS FOR THE DIAGNOSIS OF LTBI A new generation of immune-based rapid blood tests for the diagnosis of LTBI seems to be a significant upgrade of the century-old TST (11, 12). These tests possess characteristics that distinguish them from their predecessor, the Mantoux test (Table 1). They exploit the fact that the predominant host response to MTB infection consists of antigen-specific memory

Pulmonary Perspective

737

T cells releasing interferon (IFN)-␥ in response to previously encountered mycobacterial antigens. The QuantiFERON-TB Gold test (Cellestis, Carnegie, Australia), which is based on a whole-blood ELISA developed in the late 1980s, has been recently approved for in vitro diagnostics by the U.S. Food and Drug Administration, and a guideline from the U.S. Centers for Disease Control and Prevention has been published (13). A newer version of the QuantiFERON-TB Gold assay, the socalled In-Tube test, uses tubes prefilled with antigens; this might simplify the laboratory procedures, thus making the test more suitable for “on-field” usage in settings with limited resources. The T-SPOT.TB test (Oxford Immunotec, Abingdon, UK) is based on the ex vivo overnight enzyme-linked immunospot (ELISpot) assay developed by Prof. Lalvani and coworkers in the late 1990s in Oxford, United Kingdom; the group pioneered this field of clinical research, translating the results of their work to issues in clinical practice (14–18), public health (19–22), and protective immunity to TB (23–25). T-SPOT.TB has been approved for in vitro diagnostic use in Europe and is being evaluated by the U.S. Food and Drug Administration. Studies using both T-SPOT.TB and the Lalvani ELISpot assay are considered in this Pulmonary Perspective. These new tests represent the fruits of mycobacterial genomic research because they use two proteins encoded by a unique genomic segment (stretch of DNA) termed “Region of Difference 1,” which is absent from all strains of M. bovis BCG and the vast majority of nontuberculous environmental mycobacteria but is present in all clinical isolates of MTB (26). These proteins, ESAT-6 (early secretory antigenic target protein 6) and CFP10 (culture filtrate protein 10), are major targets for IFN-␥– secreting T lymphocytes in MTB-infected individuals (26, 27). Thus, the previously mentioned antigenic cross-reactivity of PPD (i.e., the main cause of poor specificity of the TST) should be, in theory, avoided by these new tests. New blood tests have an internal positive control (i.e., a sample well stimulated with a potent nonspecific stimulator of IFN-␥ production by T cells); this controls the results of the test for technical errors, including failure to add viable functioning cells to the well. Although a negative TST in immunosuppressed individuals can be a false negative, the failure of the positive

control in the blood tests provides the important information that the test’s result cannot be reliably interpreted because it may reflect an underlying in vivo immunosuppression, negatively affecting T-cell function in the in vitro stimulation. Although the two blood tests share common characteristics, they also have differences (Table 1). In the QuantiFERON-TB Gold test, a sample of whole blood with an unknown number of leukocytes is used; in T-SPOT.TB, the number of peripheral blood mononuclear cells used in the assay is quantified. The readout of the two tests is different: T-SPOT.TB enumerates individual T cells producing IFN-␥ after antigenic stimulation, and QuantiFERON-TB Gold measures the level of IFN-␥ in the supernatant of the stimulated whole blood sample. In theory, the ELISpot-based test should be more sensitive because it detects IFN-␥ in the immediate vicinity of the T cell from which it is released (where it is still at a locally high concentration), whereas the ELISA detects IFN-␥ after it has diffused into the supernatant and become diluted in the total volume of the test sample. A recent direct comparison of both tests in the same population provided initial evidence for some difference in the rate of both indeterminate and positive tests in a population including high-risk groups (28); it is possible that at least some of these differences might be explained based on the technical characteristics of the two blood assays. Another preliminary report showed that the optimal cut-off values of both tests may be slightly different from those currently recommended (29). To what extent do the blood tests affect strategies for TB control and elimination? This depends on the extent to which they overcome the known deficiencies of the TST. A critical appraisal of the performance of the new blood tests is therefore crucial. Table 2 summarizes the published evidence on sensitivity and specificity of these tests in this rapidly expanding area of clinical TB research.

SENSITIVITY OF THE BLOOD TESTS IN IMMUNOCOMPETENT INDIVIDUALS Establishing the superior diagnostic accuracy for LTBI of any test is a major challenge because there is no gold standard: however, some alternative rational approaches based on natural

TABLE 1. CHARACTERISTICS OF THE TUBERCULIN SKIN TEST AND OF THE NEW BLOOD TESTS T-SPOT.TB Antigens ESAT-6 and CFP10 Positive internal control Yes Uniformity of methods and reagents Yes Potential for boosting effect in repeated tests No Need for return visit No Time required for results 16–20 h* Setting of test In vitro Interpretation of test Objective (instrument-based) Readout units IFN-␥ spot-forming cells Technological platform ELISpot Test’s substrate PBMC Outcome measure Number of IFN-␥–producing T cells Readout system Enumeration of spots by naked eye, magnifying lens, or automated counter†

QuantiFERON-TB Gold

TST

ESAT-6 and CFP10 Yes Yes No No 16–24 h‡ In vitro Objective (instrument-based) International units of IFN-␥ ELISA Whole blood Serum concentration of IFN-␥ produced by T cells Measurement of optical density values using an automated reader

PPD No No§ Yes Yes 48–72 h|| In vivo Subjective (operator-based) Millimeters of induration NA NA NA Palpable induration

Definition of abbreviations: ELISpot ⫽ enzyme-linked ImmunoSpot; NA ⫽ not applicable; PBMC ⫽ peripheral blood mononuclear cells; PPD ⫽ protein-purified derivative; TST ⫽ tuberculin skin test. * From the T-SPOT.TB technical information sheet (www.oxfordimmunotec.com). † When enumerating the spots by the naked eye or with a magnifying glass, there is potential for more subjectivity, although a large study did show a high level of agreement between manual and automated reading of ELISpot plates (17). ‡ From the QuantiFERON-TB Gold technical information sheet (www.cellestis.com). § Mantoux test versus Heaf test, induration versus erythema, PPD-S versus PPD RT-23. || The Heaf test is read 1 week after administration.

738

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006

history and on the epidemiology of TB have been applied. The knowledge that airborne transmission of MTB is promoted by close and prolonged contact with an infectious index case (30) and by the time spent in the same room (31) has been used to generate the hypothesis that if a new test is a better marker of LTBI than the TST, it should correlate more closely with the level of exposure to MTB compared with the skin test. ELISpot has been compared with TST in five blinded, contact-tracing investigation studies. In three point-source institutional outbreaks (673 contacts in total) (19, 22, 32) and two communitybased contact tracing studies (463 contacts in total) (15, 21), ELISpot correlated with the level of exposure more strongly

than the TST. In the largest of these studies, which evaluated 535 students exposed to a single highly infectious index case at a secondary school, the difference between the two tests was statistically significant for proximity and duration of exposure (19). Although correlation with the degree of exposure to MTB is not in itself a gold standard for LTBI, these results from 1,136 TB contacts in aggregate tend to suggest that ELISpot is more sensitive than the skin test for the diagnosis of LTBI. QuantiFERON-TB Gold has also been used in one contact tracing study (125 contacts) (33) and in one case-control study (120 contacts) (34). In these studies, exposure to MTB was not measured, but contacts were allocated to two or three groups

TABLE 2. PUBLISHED STUDIES ON SENSITIVITY AND SPECIFICITY OF THE NEW BLOOD TESTS FOR THE DIAGNOSIS OF LATENT TUBERCULOSIS INFECTION T-SPOT.TB (TS.TB) Study Design Sensitivity in active TB cases HIV-negative adults

Children with high prevalence of HIV-coinfection HIV-coinfected adults

Size

QuantiFERON-TB Gold (QFT-G) Sensitivity

Prospective (38) Case-control (14) Case-control (23) Prospective (28)† Prospective (29)† Subtotal

72 47 36 24 87 266

97% 96% 92% 83% 95%

TS.TB is more sensitive than TST Blinded prospective (17)

293‡

83%

TS.TB is more sensitive than TST Case-control (16)

39

90%

Study Design Contact tracing Contact tracing Contact tracing Contact tracing Contact tracing Subtotal

Specificity in BCG-vaccinated unexposed controls

118 54 48 23 87 330

89% 81% 85% 74% 70%

No published data available for QFT-G

535 413 91 88 50 1,136

Correlation with Exposure

Study Design

Size

Contact tracing (33) Case-control (34) Subtotal

125 48 173

Correlation with Exposure Same as TST Higher than TST

Higher than TST§

TS.TB correlates with exposure to MTB better than TST

QFT-G has similar correlation with exposure to MTB as TST

Study Design

Study Design

Size

Case-control (39) Case-control (34) Subtotal

216 99 315

Case-control Case-control Case-control Case-control Subtotal

(23) (17) (16) (14)

Size 28 40 33 26 127

Specificity 100% 100% 100% 100%

TS.TB is more specific than TST Specificity in BCG-vaccinated exposed contacts

Sensitivity

No published data available for QFT-G

Size (19) (21) (32) (22) (15)

Case-control (39) Case-control (34) Prospective (47) Prospective (28)† Prospective (29)† Subtotal

Size

QFT-G is more sensitive than TST

TS.TB is highly sensitive Sensitivity in people with LTBI*

Study Design

Study Design

98% 96%

QFT-G is more specific than TST Size

Contact tracing (19)

467

Contact tracing (21) Contact tracing (32)

413 91

Contact tracing (15) Subtotal TS.TB is more specific than TST

Specificity

Association with BCG Vaccination

Study Design

Size

Contact tracing (44)

309

Subtotal

309

Association with BCG Vaccination Independent of BCG status

Independent of BCG status||

41 1,012 QFT-G is more specific than TST

Definition of abbreviations: BCG ⫽ bacille Calmette-Gue´rin; MTB ⫽ Mycobacterium tuberculosis; TB ⫽ tuberculosis; TST ⫽ tuberculin skin test. * In the absence of a gold standard for LTBI, these studies used degree of exposure to infectious index cases as a surrogate reference standard. † These studies included HIV-negative patients, but with some other immunosuppressive conditions. ‡ Of these, 30 were HIV-coinfected and with active TB. § Statistically significant (p ⫽ 0.03 and p ⫽ 0.007) difference between ELISpot and TST in (19). || Statistically significant (p ⫽ 0.001, p ⫽ 0.0001, p ⫽ 0.0003, p ⫽ 0.03, respectively) difference between ELISpot and TST.


depending on estimated risk of LTBI. In the contact tracing study on mostly BCG-unvaccinated individuals, QuantiFERONTB Gold performed similarly to TST (33). In the case-control study on mostly BCG-vaccinated subjects, in the low- and highrisk groups there were more TST-positive than QuantiFERONTB Gold–positive individuals (34). Because of the study design, it is not clear whether the high proportion of people who were positive by TST and negative by QuantiFERON-TB Gold was due to increased specificity of the blood test, decreased sensitivity with respect to TST, or a mixture of both (34). There are only two studies comparing blood and skin tests in HIV-infected individuals and infants, and these are both with ELISpot. Chapman and coworkers found that more HIVpositive adults at risk of LTBI, by virtue of residence in Zambia (a high-prevalence country), were positive by ELISpot than by TST (16). Richeldi and coworkers found that although none of 41 infants at risk of perinatal exposure to multidrug-resistant (MDR)-TB were TST positive at 6 mo of age, three were positive by ELISpot, of whom one progressed to active TB 18 mo later (22, 35). Although the TST and these new tests are not intended to be used as diagnostics for active disease, an alternative approach to the evaluation the sensitivity of new assays is to test patients with active TB. Patients with TB do not have LTBI but are infected by MTB by definition, and the results obtained could be extrapolated to LTBI. This same approach was originally used to estimate the sensitivity of TST in the early era (36). Moreover, the use of both blood tests is recommended by current guidelines as an aid for the diagnosis of active TB (13, 37). In five studies on 266 cumulative patients with TB, T-SPOT.TB showed sensitivity ranging from 83% to 97% (14, 23, 28, 29, 38). In five studies using QuantiFERON-TB Gold on 330 cumulative patients with TB, sensitivity ranged from 70% to 89% (28, 29, 34, 39) (Table 2). When these results are pooled, the sensitivity of T-SPOT.TB seems to be significantly higher compared with QuantiFERON-TB Gold in patients with active TB.

SENSITIVITY OF THE BLOOD TESTS IN IMMUNOCOMPROMISED PATIENTS A further step is to evaluate patients with TB who have risk factors for a false-negative TST (usually associated with impaired cellular immunity), who are more representative of persons with LTBI at high risk of progression (i.e., the main targets of the current control strategy). These patients usually cannot mount a proper delayed-type hypersensitivity response and are therefore potentially a challenge for blood tests based on the cellular immune response. In a study on 39 HIV-coinfected patients with TB, 92% were ELISpot positive (16). Although the patients in this study did not undergo TST, it is widely recognized that the diagnostic sensitivity of TST in this setting is around 50% (40). Furthermore, a large, prospective, blinded trial on 293 African children with suspected TB and a high rate of HIV infection and malnutrition and including many young children under 3 years of age (all factors associated with impaired cellular immunity) showed that ELISpot is significantly more sensitive than the TST in all these subgroups (17). A study on 590 HIV-positive individuals showed that the results of the QuantiFERON-TB Gold assay (In-Tube version) were closely linked with known risk factors for LTBI or past history of TB. In this study, TST results were not available, so no comparison between the blood and the skin test can be made (41); however, indeterminate results of the blood test were significantly correlated with low CD4 T-cell counts (41). A study directly comparing the two blood tests in the same population showed that QuantiFERON-TB Gold had significantly more indeterminate results and that these

739

results were significantly correlated to known risk factors for cellular immunosuppression for both tests; age below 5 years was significantly associated with indeterminate QuantiFERONTB Gold results but not with T-SPOT.TB results (28). Another study in children reported a 17% rate of indeterminate QuantiFERONTB Gold results (42); furthermore, the significant disagreement between tests among children who are likely infected suggests that the blood test may be less sensitive than the skin test for the diagnosis of LTBI (42).

SPECIFICITY OF THE BLOOD TESTS COMPARED WITH THE TST Specificity can be estimated in BCG-vaccinated individuals who are at low risk of LTBI in case-control studies, which assume that the low-risk control subjects do not have LTBI, based on the absence of epidemiologic risk factors for infection. Four casecontrol studies on 127 BCG-vaccinated subjects in aggregate showed that ELISpot has a specificity of 100% (14, 16, 20, 23, 43). A case-control study on 216 Japanese BCG-vaccinated adults showed a 98% specificity of QuantiFERON-TB Gold, much higher than for TST (35%) (39). A Korean study showed that QuantiFERON-TB Gold was positive in only 4% of 99 low-risk adults, most of them (⬎ 90%) with a BCG scar, much less than the 51% positive rate of the TST (34). Because of the lack of a gold standard test for confirming or excluding LTBI, an alternative approach to estimate specificity makes no assumptions regarding the infection status of low-risk individuals. Rather, it quantifies the correlation of test results with BCG vaccination status in people evaluated during contact investigations. These studies confirmed the higher specificity of ELISpot compared with the TST. In particular, a large contact tracing study showed that ELISpot was not associated with BCG vaccination among 535 mostly (87%) vaccinated children, whereas the TST was significantly more likely to be positive in BCG-vaccinated students (19). A recent study on 309 contacts, half of them BCG-vaccinated, confirmed that QuantiFERONTB Gold is not confounded by vaccination status (44). Overall, the evaluation of over 1,100 subjects with ELISpot and over 600 with QuantiFERON-TB Gold clearly indicated that both tests are more specific than the TST with respect to BCG vaccination status (Table 2). In summary, published studies categorically demonstrate that T-SPOT.TB and QuantiFERON-TB Gold are more specific than the TST for the diagnosis of LTBI in BCG-vaccinated populations. T-SPOT.TB, in addition, seems to be more sensitive than the TST in immunocompetent people with LTBI and in patients with active TB, including those with impaired cellular immunity at high risk of false-negative TST results. QuantiFERON-TB Gold probably has a sensitivity similar to the TST in immunocompetent people with LTBI. As with T-SPOT.TB, QuantiFERON-TB Gold has a higher sensitivity than TST in immunocompetent patients with active TB. The two blood tests have been directly compared in one study on 393 individuals, many of them affected by cellular immunosuppression; although the overall agreement between the tests was good, they performed differently in terms of positive results (28). Although these results suggesting a higher sensitivity of T-SPOT.TB have been confirmed by another comparative study on 224 individuals (29), the lack of a diagnostic gold standard for LTBI does not allow firm conclusions on a difference in sensitivity of the two blood assays.

OPERATIONAL ADVANTAGES OF THE BLOOD TESTS In addition to their improved diagnostic accuracy, the blood tests have operational advantages over TST, including lack of

740


interindividual variability in administration of the test, a more objective read-out, and a result by the next day. Two further advantages are that there is no need for a return visit to have the test read, which improves the yield of contact investigations, and the fact that repeated blood testing does not result in boosting of subsequent blood test results, which allows for repeated screening of groups at recurrent risk of TB exposure, such as health care workers. On the other hand, one significant practical drawback of the blood tests is the fact that they need to be processed within 6 h from venipuncture. Reliability of results declines after this time point, and for this reason studies using samples stored for longer than 6 h have not been included in Table 2. This operational limitation could be overcome by new formats of the tests, such as the QuantiFERON In-Tube assay; this format has been used in high-prevalence, resource-limited settings (45) and in high-risk populations (41). The improved diagnostic accuracy for the diagnosis of LTBI coupled with the advantageous operational characteristics of the new blood tests should improve the effectiveness of TB control program and will be a key factor in making TB elimination feasible in low-prevalence countries. However, achieving these goals depends on the successful deployment of the new tests under routine program conditions.

FEASIBILITY OF THE BLOOD TESTS IN ROUTINE CLINICAL PRACTICE Several reports indicate that application of the blood tests in routine clinical diagnostic microbiology laboratories and community-based contact tracing protocols is feasible. Such reports include the use of QuantiFERON-TB Gold in a routine diagnostic microbiology laboratory in an Italian University hospital (46) and in Denmark (41, 47) and the use of T-SPOT.TB in a routine diagnostic microbiology laboratory in Germany (38), in community-based contact tracing in Texas (21), and in a private diagnostic laboratory in Switzerland (48). Both tests have been used in parallel in a routine hospital setting in a microbiology laboratory in Italy (28). These technologies also seem to be feasible in relatively rudimentary laboratories in developing countries, including South Africa (17), Zambia (16), and India (45). Ferrara and coworkers noted that the QuantiFERON-TB Gold positive control failed in 21% of tests (n ⫽ 318) and that these indeterminate results were significantly overrepresented among patients receiving immunosuppressive treatments (46). A similar result has been confirmed in a prospective study on 393 individuals (28). A positive-control failure rate of 12% was also observed in a Danish study (47), and 17% of blood tests failed among children evaluated in Australia (42). Indeterminate results due to positive control failures seem to be less frequent with T-SPOT.TB (28, 38). This also applies to the use of T-SPOT.TB in subgroups with impaired cellular immunity, with 0 of 41 neonates exposed to MDR-TB and 0 of 60 HIV-infected adults with active TB or LTBI in Zambia giving failed positive controls (16, 22). A recent report of 29 HIV-infected subjects tested with T-SPOT.TB gave only one (3%) indeterminate result, and positive control responses were not adversely affected by low CD4 counts (49). On the other hand, a larger study reported a high rate (24%) of indeterminate QuantiFERONTB Gold In-Tube results among HIV-positive patients with fewer than 100 CD4 T cells/␮l (41). Finally, three case reports provided interesting initial evidence of the clinical utility of these tests in supporting a diagnosis of active TB in TST-negative immunosuppressed patients. ELISpot results prompted early diagnosis of active MDR-TB in a skinnegative asymptomatic contact undergoing long-term therapy with azathioprine for Crohn disease (18). ELISpot was the only

positive test in a patient with hairy cell leukemia and disseminated active TB (50). QuantiFERON-TB Gold was positive in the context of a false-negative TST in a patient with polymyositis undergoing immunosuppressive treatment who subsequently developed active TB (51).

IMPACT OF THE BLOOD TESTS ON TARGETED TESTING OF LTBI What will be the societal impact and cost to health care systems when future guidelines incorporate the new blood tests? Replacement of the skin test with the blood tests will not change the principles of targeted testing, which are based on identifying those groups at highest risk of progression to active TB (1). Therefore, the advent of the new blood tests should not lead to global, indiscriminate population screening. Higher specificity will reduce or eliminate false-positive test results in BCG-vaccinated people, thus avoiding the costs associated with unnecessary chemoprophylaxis and its associated toxicity. Higher sensitivity would, on the other hand, identify more infected persons among those with a false-negative TST result. More true-positive results in infected people would increase the rate of diagnosis and treatment of LTBI in the most vulnerable populations before progression to active TB. The introduction of blood tests in clinical practice would initially increase the cost of TB control. The fact that the majority of costs for TB control are incurred during diagnosis and treatment of inpatients with active TB suggests that higher diagnostic sensitivity and higher specificity could generate cost savings in the medium and long term by reducing the future burden of cases of active TB and decreasing the number of uninfected BCG-vaccinated people inappropriately treated for LTBI. However, the new blood tests will transfer a workload to diagnostic laboratories, which are usually not in charge of performing the skin test; moreover, T-SPOT.TB requires some technical skills and dedicated equipment. Selecting which test to use among the two blood tests or between a blood and the skin test in a given epidemiologic situation depends on the population being tested, the purpose of testing, and the resources available. In principle, direct implementation and full replacement of TST with blood tests would be the most accurate and simplest strategy because only one system would be in place and overall performance of targeted testing would be improved. This strategy has been suggested by the Centers for Disease Control and Prevention in their guideline on QuantiFERON-TB Gold (13). More data in the most vulnerable groups (such as HIV-infected persons and young children) are needed before the tests may enter clinical practice and fully replace the current diagnostic standard. Furthermore, it is predictable that the skin test will be in use for some time in parts of the world where the new blood tests may not be affordable. A two-step approach based on the use of a blood test as a confirmatory test in all those with positive TST results and a primary test in all individuals with risk factors for false-negative TST results has recently been proposed in the new U.K. national guideline on TB (37). This approach is being considered on the basis of economic modeling and was supported because of clinical utility and feasibility. However, because false-negative TST results are not fully predictable based on individuals’ characteristics (e.g., in immigrants with unrecognized HIV infection), some of the subjects at higher risk of progression would not be identified using this strategy. A two-step protocol would require two diagnostic systems (blood test and skin test) and their requisite expertise to be actively maintained in parallel. This would result in an increase in the complexity of the diagnostic algorithm for LTBI and number of return visits required. Also, one would


have to consider the potential effect of previous in vivo tuberculin testing on subsequent results with the ex vivo blood tests. PPD contains ESAT-6 and CFP10 and might, at least in theory, induce false-positive blood test results by inducing T-cell responses to these antigens. In this regard, it is reassuring that ELISpot results were not affected by three previous serial TSTs in a longitudinal study of 42 people (52). Initially, the cost of a blood test in Europe and in the United States should be 25–35 Euros (or about $30–40). These costs are higher compared with the net cost of the TST, but it is reasonable to predict that they will decrease with the increase of usage of the tests. Moreover, these higher costs need to be evaluated in the context of a potential significant increase in the efficacy of TB control strategies. Two recent health economic studies show that using the blood test to detect truly infected individuals is cost-effective and may positively affect control strategies (53, 54). Further research in certain areas is required. Although the published data suggest that T-SPOT.TB is more sensitive than QuantiFERON-TB Gold in LTBI and active TB, the absence of a gold standard prevents any definitive conclusion; more headto-head comparative studies of the two tests are required. Another research priority is the evaluation of both tests for the diagnosis of LTBI in patients with iatrogenic immunosuppression (e.g., dialysis, organ transplantation, and anti–tumor necrosis factor-␣ treatment). Finally, longitudinal studies of these and other high-risk cohorts will enable evaluation of the risk of progression in individuals with positive blood test results; this will help to define the positive predictive value for subsequent development of active TB for both blood tests. The last decade of basic science research into TB has generated much new important information about the bacillus and the host response, but this has yet to result in any significant improvement in how we diagnose and manage our patients and their contacts. It is likely that the introduction of the new blood tests for the diagnosis of LTBI into TB control and elimination strategies will represent the first significant improvement to arise from basic science research since the discovery of rifampin in Italy half a century ago. Conflict of Interest Statement : L.R. received €1,900 in 2005 for serving on an advisory board for Oxford Immunotec Ltd., manufacturer of T-SPOT.TB. His institution received an unrestricted grant (€6,000) from A.D.A. SRL, representative of Cellestis in Italy for the QuantiFERON-TB Gold Test. Acknowledgment : The author thanks M. Losi and G. Ferrara (University of Modena and Reggio Emilia, Modena, Italy) for their critical appraisal of the manuscript.

References 1. American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med 2000;161: S221–S247. 2. Ferebee SH. Controlled chemoprophylazis trials in tuberculosis: a general review. Adv Tuberc Res 1969;17:29–106. 3. Horsburgh CR Jr. Priorities for the treatment of latent tuberculosis infection in the United States. N Engl J Med 2004;350:2060–2067. 4. Huebner RE, Schein MF, Bass JB Jr. The tuberculin skin test. Clin Infect Dis 1993;17:968–975. 5. American Thoracic Society. Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 2000; 161:1376–1395. 6. Seibert FB, Glenn JT. Tuberculin purified protein derivative: preparation and analyses of a large quantity for standard. Am Rev Tuberc 1944;44:9. 7. Andersen P, Doherty TM. The success and failure of BCG: implications for a novel tuberculosis vaccine. Nat Rev Microbiol 2005;3:656–662. 8. Tissot F, Zanetti G, Francioli P, Zellweger JP, Zysset F. Influence of bacille Calmette-Guerin vaccination on size of tuberculin skin test reaction: to what size? Clin Infect Dis 2005;40:211–217. 9. Wang L, Turner MO, Elwood RK, Schulzer M, FitzGerald JM. A metaanalysis of the effect of Bacille Calmette Guerin vaccination on tuberculin skin test measurements. Thorax 2002;57:804–809.

741 10. Joint Tuberculosis Committee of the British Thoracic Society. Control and prevention of tuberculosis in the UK: code of practice 2000. Thorax 2000;55:887–901. 11. Barnes PF. Diagnosing latent tuberculosis infection: the 100-year upgrade. Am J Respir Crit Care Med 2001;163:807–808. 12. Barnes PF. Diagnosing latent tuberculosis infection: turning glitter to gold. Am J Respir Crit Care Med 2004;170:5–6. 13. Mazurek GH, Jereb J, Lobue P, Iademarco MF, Metchock B, Vernon A. Guidelines for using the QuantiFERON-TB Gold test for detecting Mycobacterium tuberculosis infection, United States. MMWR Recomm Rep 2005;54:49–55. 14. Lalvani A, Pathan AA, McShane H, Wilkinson RJ, Latif M, Conlon CP, Pasvol G, Hill AV. Rapid detection of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Am J Respir Crit Care Med 2001;163:824–828. 15. Lalvani A, Pathan AA, Durkan H, Wilkinson KA, Whelan A, Deeks JJ, Reece WH, Latif M, Pasvol G, Hill AV. Enhanced contact tracing and spatial tracking of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Lancet 2001;357:2017–2021. 16. Chapman AL, Munkanta M, Wilkinson KA, Pathan AA, Ewer K, Ayles H, Reece WH, Mwinga A, Godfrey-Faussett P, Lalvani A. Rapid detection of active and latent tuberculosis infection in HIV-positive individuals by enumeration of Mycobacterium tuberculosis-specific T cells. AIDS 2002;16:2285–2293. 17. Liebeschuetz S, Bamber S, Ewer K, Deeks J, Pathan AA, Lalvani A. Diagnosis of tuberculosis in South African children with a T-cell-based assay: a prospective cohort study. Lancet 2004;364:2196–2203. 18. Richeldi L, Ewer K, Losi M, Hansell DM, Roversi P, Fabbri LM, Lalvani A. Early diagnosis of subclinical multidrug-resistant tuberculosis. Ann Intern Med 2004;140:709–713. 19. Ewer K, Deeks J, Alvarez L, Bryant G, Waller S, Andersen P, Monk P, Lalvani A. Comparison of T-cell-based assay with tuberculin skin test for diagnosis of Mycobacterium tuberculosis infection in a school tuberculosis outbreak. Lancet 2003;361:1168–1173. 20. Lalvani A, Nagvenkar P, Udwadia Z, Pathan AA, Wilkinson KA, Shastri JS, Ewer K, Hill AV, Mehta A, Rodrigues C. Enumeration of T cells specific for RD1-encoded antigens suggests a high prevalence of latent Mycobacterium tuberculosis infection in healthy urban Indians. J Infect Dis 2001;183:469–477. 21. Shams H, Weis SE, Klucar P, Lalvani A, Moonan PK, Pogoda JM, Ewer K, Barnes PF. Enzyme-linked immunospot and tuberculin skin testing to detect latent tuberculosis infection. Am J Respir Crit Care Med 2005;172:1161–1168. 22. Richeldi L, Ewer K, Losi M, Bergamini BM, Roversi P, Deeks J, Fabbri LM, Lalvani A. T cell-based tracking of multidrug resistant tuberculosis infection after brief exposure. Am J Respir Crit Care Med 2004;170: 288–295. 23. Pathan AA, Wilkinson KA, Klenerman P, McShane H, Davidson RN, Pasvol G, Hill AV, Lalvani A. Direct ex vivo analysis of antigen-specific IFN-gamma-secreting CD4 T cells in Mycobacterium tuberculosisinfected individuals: associations with clinical disease state and effect of treatment. J Immunol 2001;167:5217–5225. 24. Pathan AA, Wilkinson KA, Wilkinson RJ, Latif M, McShane H, Pasvol G, Hill AV, Lalvani A. High frequencies of circulating IFN-gammasecreting CD8 cytotoxic T cells specific for a novel MHC class Irestricted Mycobacterium tuberculosis epitope in M. tuberculosisinfected subjects without disease. Eur J Immunol 2000;30:2713–2721. 25. Soysal A, Millington KA, Bakir M, Dosanjh D, Aslan Y, Deeks JJ, Efe S, Staveley I, Ewer K, Lalvani A. Effect of BCG vaccination on risk of Mycobacterium tuberculosis infection in children with household tuberculosis contact: a prospective community-based study. Lancet 2005;366:1443–1451. 26. Andersen P, Munk ME, Pollock JM, Doherty TM. Specific immunebased diagnosis of tuberculosis. Lancet 2000;356:1099–1104. 27. Tully G, Kortsik C, Hohn H, Zehbe I, Hitzler WE, Neukirch C, Freitag K, Kayser K, Maeurer MJ. Highly focused T cell responses in latent human pulmonary Mycobacterium tuberculosis infection. J Immunol 2005;174:2174–2184. 28. Ferrara G, Losi M, D’Amico R, Roversi P, Piro R, Meacci M, Meccugni B, Marchetti Dori I, Andreani A, Bergamini B, et al. Use in routine clinical practice of two commercial blood tests for diagnosis of infection with Mycobacterium tuberculosis: a prospective study. Lancet 2006; 367:1328–1334. 29. Lee JY, Choi HJ, Park IN, Hong SB, Oh YM, Lim CM, Lee SD, Koh Y, Kim WS, Kim DS, Kim WD, Shim TS. Comparison of two commercial interferon gamma assays for diagnosing Mycobacterium tuberculosis infection. Eur Respir J 2006;28:24–30.

742


30. Kenyon TA, Valway SE, Ihle WW, Onorato IM, Castro KG. Transmission of multidrug-resistant Mycobacterium tuberculosis during a long airplane flight. N Engl J Med 1996;334:933–938. 31. Houk VN, Baker JH, Sorensen K, Kent DC. The epidemiology of tuberculosis infection in a closed environment. Arch Environ Health 1968;16: 26–35. 32. Zellweger JP, Zellweger A, Ansermet S, de Senarclens B, WrightonSmith P. Contact tracing using a new T-cell-based test: better correlation with tuberculosis exposure than the tuberculin skin test. Int J Tuberc Lung Dis 2005;9:1242–1247. 33. Brock I, Weldingh K, Lillebaek T, Follmann F, Andersen P. Comparison of tuberculin skin test and new specific blood test in tuberculosis contacts. Am J Respir Crit Care Med 2004;170:65–69. 34. Kang YA, Lee HW, Yoon HI, Cho B, Han SK, Shim YS, Yim JJ. Discrepancy between the tuberculin skin test and the whole-blood interferon gamma assay for the diagnosis of latent tuberculosis infection in an intermediate tuberculosis-burden country. JAMA 2005;293: 2756–2761. 35. Richeldi L, Ewer K, Losi M, Bergamini B, Millington KA, Fabbri LM, Lalvani A. T cell-based diagnosis of neonatal multidrug-resistant latent tuberculosis infection. Pediatrics (In press) 36. Edwards LB, Acquaviva FA, Livesay VT. Identification of tuberculosis infected. Am Rev Respir Dis 1973;108:1334–1339. 37. National Collaborating Centre for Chronic Conditions. Tuberculosis: clinical diagnosis and management of tuberculosis, and measures for its prevention and control. London: Royal College of Physicians; 2006. 38. Meier T, Eulenbruch HP, Wrighton-Smith P, Enders G, Regnath T. Sensitivity of a new commercial enzyme-linked immunospot assay (T SPOT-TB) for diagnosis of tuberculosis in clinical practice. Eur J Clin Microbiol Infect Dis 2005;24:529–536. 39. Mori T, Sakatani M, Yamagishi F, Takashima T, Kawabe Y, Nagao K, Shigeto E, Harada N, Mitarai S, Okada M, et al. Specific detection of tuberculosis infection: an interferon-gamma-based assay using new antigens. Am J Respir Crit Care Med 2004;170:59–64. 40. Shafer RW, Singh SP, Larkin C, Small PM. Exogenous reinfection with multidrug-resistant Mycobacterium tuberculosis in an immunocompetent patient. Tuber Lung Dis 1995;76:575–577. 41. Brock I, Ruhwald M, Lundgren B, Westh H, Mathiesen LR, Ravn P. Latent tuberculosis in HIV positive, diagnosed by the M. Tuberculosis Specific Interferon Gamma test. Respir Res 2006;7:56. 42. Connell TG, Curtis N, Ranganathan SC, Buttery JP. Performance of a whole blood interferon gamma assay for detecting latent infection with Mycobacterium tuberculosis in children. Thorax 2006;61:616–620. 43. Liu XQ, Dosanjh D, Varia H, Ewer K, Cockle P, Pasvol G, Lalvani

44.

45.

46.

47.

48.

49.

50.

51.

52.

53. 54.

A. Evaluation of T-cell responses to novel RD1- and RD2-encoded Mycobacterium tuberculosis gene products for specific detection of human tuberculosis infection. Infect Immun 2004;72:2574–2581. Diel R, Nienhaus A, Lange C, Meywald-Walter K, Forssbohm M, Schaberg T. Tuberculosis contact investigation with a new, specific blood test in a low-incidence population containing a high proportion of BCGvaccinated persons. Respir Res 2006;7:77. Pai M, Gokhale K, Joshi R, Dogra S, Kalantri S, Mendiratta DK, Narang P, Daley CL, Granich RM, Mazurek GH, et al. Mycobacterium tuberculosis infection in health care workers in rural India: comparison of a whole-blood interferon gamma assay with tuberculin skin testing. JAMA 2005;293:2746–2755. Ferrara G, Losi M, Meacci M, Meccugni B, Piro R, Roversi P, Bergamini BM, D’Amico R, Marchegiano P, Rumpianesi F, et al. Routine hospital use of a new commercial whole blood interferon-gamma assay for the diagnosis of tuberculosis infection. Am J Respir Crit Care Med 2005;172:631–635. Ravn P, Munk ME, Andersen AB, Lundgren B, Lundgren JD, Nielsen LN, Kok-Jensen A, Andersen P, Weldingh K. Prospective evaluation of a whole-blood test using Mycobacterium tuberculosis-specific antigens ESAT-6 and CFP-10 for diagnosis of active tuberculosis. Clin Diagn Lab Immunol 2005;12:491–496. Zellweger JP, Zellweger A, Ansermet S, de Senarclens B, WrightonSmith P. New T cell based test correlates better with tuberculosis exposure than tuberculin skin test. Int J Tuberc Lung Dis 2005;9:1242– 1247. Dheda K, Lalvani A, Miller RF, Scott G, Booth H, Johnson MA, Zumla A, Rook GA. Performance of a T-cell-based diagnostic test for tuberculosis infection in HIV-infected individuals is independent of CD4 cell count. AIDS 2005;19:2038–2041. Richeldi L, Luppi M, Losi M, Luppi F, Potenza L, Roversi P, Cerri S, Millington KA, Ewer K, Fabbri LM, et al. Diagnosis of occult tuberculosis in hematological malignancy by enumeration of antigen-specific T cells. Leukemia 2006;20:379–381. Ravn P, Munk ME, Andersen AB, Lundgren B, Nielsen LN, Lillebaek T, Soerensen IJ, Andersen P, Weldingh K. Reactivation of tuberculosis during immunosuppressive treatment in a patient with a positive QuantiFERON-RD1 test. Scand J Infect Dis 2004;36:499–501. Richeldi L, Ewer K, Losi M, Roversi P, Fabbri LM, Lalvani A. Repeated tuberculin testing does not induce false positive ELISpot results. Thorax 2005;61:180. Wrighton-Smith P, Zellweger JP. Direct costs of three models for the screening of latent tuberculosis infection. Eur Respir J 2006;28:45–50. Diel R, Nienhaus A, Lange C, Schaberg T. Cost optimization of screening for latent tuberculosis in close contacts. Eur Respir J 2006;28:35–44.