Untitled

C L I N I C A L A N D L A B O R A T O R Y I N V E S T I G A T I O N S Tuberculosis screening in patients with psoriasis beforeantitumour necrosis factor therapy: comparison of aninterferon-c release assay vs. tuberculin skin testE. Laffitte, J.P. Janssens,* P. Roux-Lombard,  A.M. Thielen, C. Barde, G. Marazza, R.G. Panizzonà andJ.-H. Saurat Clinic of Dermatology, and Divisions of *Pulmonary Diseases and  Immunology and Allergy, University Medical Hospital, CH-1211 Geneva, SwitzerlandàClinic of Dermatology, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland Background Antitumour necrosis factor (anti-TNF) treatments may reactivate latent tuberculosis infection (LTBI). For detecting LTBI, the tuberculin skin test (TST) has low sensitivity and specificity. Interferon-c release assays (IGRA) have beenshown to be more sensitive and specific than TST.
Objective To compare the TST and the T-SPOT.TB IGRA for identifying LTBI in patients with psoriasis before anti-TNF treatment.
Methods A retrospective study was carried out over a 4-year period on patients adalimumab, etanercept, infliximab, psoriasis, with psoriasis requiring anti-TNF treatment. All were subjected to the TST, T-SPOT.TB and chest X-ray. Risk factors for LTBI and history of bacillusCalmette–Gue´rin (BCG) vaccination were recorded. The association of T-SPOT.TB Conflicts of interestNone declared.
and TST results with risk factors for LTBI was tested through univariate logisticregression models. Agreement between tests was quantified using kappa statistics.
Treatment for LTBI was started 1 month before anti-TNF therapy when indicated.
Congress Psoriasis from Gene to Clinic, London, Results Fifty patients were included; 90% had prior BCG vaccination. A positive U.K., 4–6 December 2008 and awarded the T-SPOT.TB was strongly associated with a presumptive diagnosis of LTBI (odds ratio 7Æ43; 95% confidence interval 1Æ38–39Æ9), which was not the case for the TST. Agreement between the T-SPOT.TB and TST was poor, j = 0Æ33 (SD 0Æ13).
LTBI was detected and treated in 20% of the patients. In 20% of the cases, LTBIwas not retained in spite of a positive TST but a negative T-SPOT.TB. All patientsreceived an anti-TNF agent for a median of 56 weeks (range 20–188); amongpatients with a positive TST ⁄negative T-SPOT.TB, no tuberculosis was detectedwith a median follow-up of 64 weeks (44–188). One case of disseminatedtuberculosis occurred after 28 weeks of adalimumab treatment in a patient withLTBI in spite of treatment with rifampicin.
Conclusion This study is the first to underline the frequency of LTBI in patientswith psoriasis (20%), and to support the use of IGRA instead of the TST for itsdetection. Nevertheless, there is still a risk of tuberculosis under anti-TNF ther-apy, even if LTBI is correctly diagnosed and treated.
Antitumour necrosis factor (anti-TNF)-a agents are approved ity with false-positive results in bacillus Calmette–Gue´rin for the treatment of psoriasis or other inflammatory diseases.
(BCG)-vaccinated subjects. This leads to unnecessary treat- However, they may reactivate latent tuberculosis infection ments for LTBI with a significant risk of drug toxicity, and (LTBI).1,2 Thus, screening for LTBI is mandatory and preven- lower sensitivity in immunosuppressed patients compared tive treatment should be given to all patients with evidence of with healthy subjects resulting in false-negative results and a LTBI before starting any anti-TNF-a therapy,3 even if it does subsequent risk of tuberculosis reactivation with anti-TNF not offer complete protection.4 Screening for LTBI is tradition- therapy. Finally, the limit above which the TST is considered ally based on history, chest X-ray and the tuberculin skin test positive (i.e. indicative of latent infection) differs according to (TST).5–7 However, the TST has disadvantages: a low specific- countries and guidelines (5–10 mm).
Journal Compilation Ó 2009 British Association of Dermatologists • British Journal of Dermatology 2009 2 Tuberculosis screening in psoriasis patients, E. Laffitte et al.
Since the early 2000s, in vitro blood tests measuring produc- univariate logistic regression models. Agreement between tests tion of interferon (IFN)-c by T cells exposed to antigens was quantified using kappa statistics. All statistical analyses highly specific for Mycobacterium tuberculosis have been developed.
were performed with SPSS version 14.0, 2006 (Statistical Pack- These tests (T-SPOT.TB; Oxford Immunotec, Oxford, U.K. and age for Social Sciences Inc., Chicago, IL, U.S.A.).
QuantiFERONÒ-TB Gold; Cellestis, Carnegie, Australia), collec- If LTBI was diagnosed, treatment with either rifampicin tively referred to as IFN-c release assays (IGRAs), are not (10 mg kg)1, max. 600 mg daily) for 4 months or isoniazid affected by prior BCG vaccination, thus offering increased (5 mg kg)1, max. 300 mg daily) for 9 months was started specificity compared with the TST.8 Their sensitivity, extrapo- 1 month before introducing anti-TNF-a therapy, according to lated from studies in patients with active tuberculosis, is prob- ably at least as good as that of the TST in the detection ofLTBI. Recent guidelines have integrated the use of IGRAs in screening strategies for subjects exposed to tuberculosis. TheU.S. Centers for Disease Control and Prevention guidelines9 Fifty patients were analysed (Table 1). Ten patients (20%) came state that the QuantiFERON-TB Gold test can be used in all from or had lived in a country with a high incidence of tuber- circumstances in which the TST is used and can be used in culosis. Forty-five patients (90%) had prior BCG vaccination.
place of the TST. The U.K. National Institute for Health andClinical Excellence 2006 guidelines10 and the 2007 Swiss Comparative results of tuberculin skin test and national guidelines11 both recommend, for immunocompetent adults, a two-step procedure, i.e. confirmation of positive TSTresults using an IGRA.
In 28 patients (56%), the TST and T-SPOT.TB were both nega- However, there are few data on LTBI detection and use of tive. In eight patients (16%) both the T-SPOT.TB and TST IGRAs in patients with psoriasis. The aim of this study was (‡ 5 mm) were positive. Two cases had a negative TST but a (i) to determine the frequency of LTBI in a population of positive T-SPOT.TB with a chest X-ray suggestive of LTBI (gran- patients with psoriasis before anti-TNF treatment, (ii) to com- uloma). In 12 cases (24%) the TST was positive and the pare the TST with T-SPOT.TB for detecting LTBI, and (iii) to T-SPOT.TB was negative; all had prior BCG vaccination and a evaluate the tolerance and effectiveness of treatment for LTBI normal chest X-ray. Agreement between both tests was poor for under anti-TNF therapy in our patients.
TST ‡ 5 vs. T-SPOT.TB: 36 ⁄ 50 (72%), j = 0Æ33 (SD 0Æ13).
Probability of having a positive interferon-c releaseassay or tuberculin skin test according to specific risk This retrospective study was conducted in two academic der- matological centres (University Hospital, Geneva and CHUV,Lausanne). All patients seen between November 2004 and Table 2 shows the association between the presence of risk March 2008 with moderate to severe psoriasis qualifying for factors for LTBI and results of an IGRA and TST based on uni- anti-TNF-a therapy were screened for LTBI with a chest X-ray, variate logistic regression analyses. Significant associations a TST and a T-SPOT.TB IGRA. The TST was considered positive were found between having a positive T-SPOT.TB and proba- if the induration diameter was ‡ 5 mm. As there is no ‘gold ble LTBI, a chest X-ray suggestive of LTBI or a history of standard’ test for the diagnosis of LTBI, we aimed to evaluate exposure to active tuberculosis. There was no association with the risk factors for LTBI by recording the following data: age; age, previous immunosuppressive therapy or country of origin country of origin and tuberculosis incidence in the country of with a high incidence of tuberculosis. Conversely, none of the origin according to World Health Organization data; history above-mentioned variables was associated with having a TST of tuberculosis exposure (family or work); history of pro- either ‡ 5 or 10. Association with BCG-vaccination status longed stay in a high-incidence area; BCG vaccination; and could not be tested, because of the high rate of patients vacci- prior immunosuppressive therapy (methotrexate, ciclosporin nated (90%). Results from the multivariable analysis (not In the absence of a ‘gold standard’ for the diagnosis of LTBI, we analysed results of the T-SPOT.TB and the TST in Latent tuberculosis infection diagnosis and treatment relation to the risk factors for LTBI, BCG-vaccination status,and to a composite variable defining a probable diagnosis of A diagnosis of LTBI was considered in the 10 cases (20%) LTBI (‘probable LTBI’) defined as having a history of definite with a positive T-SPOT.TB. In 28 patients with both a negative exposure to a case of active tuberculosis and ⁄ or having a chest TST and T-SPOT.TB, LTBI was considered as reasonably X-ray suggestive of prior tuberculosis infection (granulomas, excluded. In 12 cases (24%) with a TST ‡ 5 but a negative calcified adenopathy) and ⁄ or originating from a high-inci- T-SPOT.TB, all with a prior BCG vaccination and a normal dence country (defined as > 40 cases in 100 000 per year).
X-ray, a diagnosis of LTBI was not considered.
The association of either T-SPOT.TB or TST results with risk A total of 12 patients were treated for LTBI (nine with rif- factors for LTBI or BCG-vaccination status was tested through ampicin and three with isoniazid): the 10 patients with LTBI, Journal Compilation Ó 2009 British Association of Dermatologists • British Journal of Dermatology 2009 Tuberculosis screening in psoriasis patients, E. Laffitte et al. 3 Table 1 Characteristics of the 50 patients analysed 54Æ3 (25–63) 46Æ4 (17–75) 47Æ5 (17–64) 48Æ3 (25–74) 49Æ2 (25–64) 47Æ3 (17–75) BCG, bacillus Calmette–Gue´rin; LTBI, latent tuberculosis infection; TST, tuberculin skin test; Pos, positive; Neg, negative. aMethotrexate,ciclosporin or efalizumab. bProbable LTBI was defined as having a history of definite exposure to a case of active TB and ⁄ or having a chestX-ray suggestive of prior TB infection (granulomas, calcified adenopathy) and ⁄ or originating from a country with a high incidence (definedas > 40 in 100 000 per year).
Table 2 Association of either T-SPOT.TB or the tuberculin skin test (TST) results with risk factors for latent tuberculosis infection (LTBI):univariate logistic regression models CI, confidence interval. aPrior immunosuppressive therapy: methotrexate, ciclosporin or efalizumab. bProbable LTBI defined as having ahistory of definite exposure to a case of active TB and ⁄ or having a chest X-ray suggestive of prior TB infection (granulomas, calcifiedadenopathy) and ⁄ or originating from a country with a high incidence (defined as > 40 in 100 000 per year). Bold text: significant associa-tion (P < 0Æ01) and two patients with a positive TST (‡ 5 mm) and a negative of adalimumab in a patient with LTBI treated with rifampicin T-SPOT.TB but included in a clinical trial which did not accept the validity of T-SPOT.TB. There were no serious adverseeffects with the LTBI treatment; two patients had an elevation of transaminase less than five times the normal limit, whichdid not required the interruption of therapy.
The present study is to our knowledge the first to show that,in a population of patients treated for psoriasis, a positiveT-SPOT.TB IGRA is strongly associated with the presence of risk factors for LTBI. This association was not found for the Patients received anti-TNF-a therapy (etanercept, infliximab or TST, and agreement between the T-SPOT.TB and TST was adalimumab) for a median of 56 weeks (range 20–188). No poor, probably because of a high rate of BCG-vaccinated case of tuberculosis occurred in the 10 patients with a positive patients (90%) acting as a confounding factor. As there is no TST ⁄ negative T-SPOT.TB who were not treated for LTBI, after gold standard for the diagnosis of LTBI, it was impossible to a median duration of 64 weeks of anti-TNF therapy (range estimate the true sensitivity and specificity of either test in this 44–188). A disseminated tuberculosis occurred after 28 weeks Journal Compilation Ó 2009 British Association of Dermatologists • British Journal of Dermatology 2009 4 Tuberculosis screening in psoriasis patients, E. Laffitte et al.
There are very few data concerning LTBI and the use of to show in this population highly significant univariate results IGRAs in patients with psoriasis. In a recent report of 11 between risk factors for LTBI and the T-SPOT.TB, and support patients, Desai et al.12 studied the utility of another IGRA the use of IGRAs instead of the TST.
(QuantiFERON) in screening for LTBI before anti-TNF ther-apy. They suggest that QuantiFERON should replace the TST for LTBI screening because its validity is well documented in 1 Keane J, Bresnihan B. Tuberculosis reactivation during immuno- rheumatological disorders, but there was no evaluation of risk suppressive therapy in rheumatic diseases: diagnostic and therapeu- factors for LTBI in the patients assessed.
tic strategies. Curr Opin Rheumatol 2008; 20:443–9.
We chose to base our diagnosis of LTBI on the results of 2 Gardam MA, Keystone EC, Menzies R et al. Antitumour necrosis the T-SPOT.TB rather than on the TST. Several points suggest factor agents and tuberculosis risk: mechanisms of action and clini- that patients with LTBI were correctly identified: patients with cal management. Lancet Infect Dis 2003; 3:148–55.
a positive T-SPOT.TB had a significant association with risk 3 Lin J, Ziring D, Desai S et al. TNFalpha blockade in human diseases: factors for LTBI; and in 10 of 12 patients with a positive TST an overview of efficacy and safety. Clin Immunol 2008; 126:13–30.
4 Sichletidis L, Settas L, Spyratos D et al. Tuberculosis in patients and a negative T-SPOT.TB, no treatment was given and no receiving anti-TNF agents despite chemoprophylaxis. Int J Tuberc tuberculosis occurred under anti-TNF therapy with a mean follow-up of 76 weeks. According to the literature, most cases 5 British Thoracic Society Standards of Care Committee. BTS recom- of tuberculosis in patients receiving anti-TNF therapy are mendations for assessing risk and for managing Mycobacterium tubercu- detected within 12 months of the beginning of anti-TNF treat- losis infection and disease in patients due to start anti-TNF-alpha ment.13 Treatment for LTBI was administered in 12 cases, and was avoided in 10 patients, whereas it would have been given 6 Doherty SD, Van Voorhees A, Lebwohl MG et al. National Psoriasis Foundation consensus statement on screening for latent tuberculo- to 22 patients if we have followed the guidelines based on the sis infection in patients with psoriasis treated with systemic and TST alone. Treatment for LTBI was relatively well tolerated, biologic agents. J Am Acad Dermatol 2008; 59:209–17.
with a slight elevation of transaminases in two patients.
7 Centers for Disease Control and Prevention. Guide for primary health Twenty per cent of our patients had a probable LTBI which care providers: targeted tuberculin testing and treatment of latent tuberculosis infec- we consider to be a high rate; however, we were unable to tion. Available at: http://www.cdc.gov/tb/publications/LTBI/pdf/ compare our results with those from other studies, as pub- TargetedLTBI05.pdf (last accessed 10 June 2009).
lished data only report the frequency of tuberculosis in 8 Menzies D, Pai M, Comstock G. Meta-analysis: new tests for the diagnosis of latent tuberculosis infection: areas of uncertainty and patients under an anti-TNF agent for moderate to severe recommendations for research. Ann Intern Med 2007; 146:340–54.
plaque psoriasis.6 In these studies, the patients were screened 9 Mazurek GH, Jereb J, Lobue P et al. Guidelines for using the for LTBI and treated if necessary before the anti-TNF treat- QuantiFERON-TB Gold test for detecting Mycobacterium tuberculosis ment, but the actual number diagnosed with LTBI, and the infection, United States. MMWR Recomm Rep 2005; 54:49–55.
tolerance of its therapy are not reported.
10 National Institute for Health and Clinical Excellence. Tuberculosis.
An important message from our observations is that the risk Clinical Diagnosis and Management of tuberculosis, and measures for of active tuberculosis under anti-TNF therapy persists even if its prevention and control. Available at: http://www.nice.org.uk/CG033. Last updated March 2006 (last accessed 10 June 2009).
LTBI is diagnosed and treated. Indeed, a case of disseminated 11 Beglinger C, Dudler J, Mottet C et al. Screening for tuberculosis tuberculosis occurred in a patient adequately treated for LTBI.
infection before the initiation of an anti-TNF-alpha therapy. Swiss Even if screening of LTBI reduces the occurrence of active tuberculosis under anti-TNF therapy,14 a residual risk remains, 12 Desai N, Raste Y, Cooke NT, Harland CC. QuantiFERON-TB Gold with atypical and disseminated presentations that dermatolo- testing for tuberculosis in psoriasis patients commencing anti- Our study had a few limitations. It was retrospective in 13 Bieber J, Kavanaugh A. Tuberculosis and opportunistic infections: design, but as all patients seen in both institutions were analy- relevance to biologic agents. Clin Exp Rheumatol 2004; 22:S126–33.
sed, the risk of selection bias was limited. Also, we included a 14 Gomez-Reino JJ, Carmona L, Angel Descalzo M. Risk of tuberculo- relatively low number of patients, which may have limited the sis in patients treated with tumor necrosis factor antagonists due to feasibility of certain statistical computations (i.e. multivariate incomplete prevention of reactivation of latent infection. Arthritis logistic regression). However, the data presented are the first Journal Compilation Ó 2009 British Association of Dermatologists • British Journal of Dermatology 2009

Source: http://www.redetb.org.br/attachments/549_PUB_ART_Laffitte_BJD_2009.pdf