Quantiferon-TB Gold test in screening for latent tuberculosis before and during antitumour necrosis factor treatment

All candidates for anti-TNF treatment should undergo screening for latent tuberculosis infection (LTBI). BCG vaccination may lead to a positive tuberculin skin test (TST) and false diagnosis of LTBI. Conversely, some immunosuppressed patients may not respond to a TST. Quantiferon TB-Gold (QFT-G) is a new screening tool that uses a peptide cocktail simulating ESAT-6, CFP-10 and TB7.7(p4) proteins to stimulate cells in heparinised whole blood. Detection of IFNγ by ELISA is used to identify in vitro responses to these peptide antigens that are specifically associated with Mycobacterium tuberculosis infection and do not cross-react with immunity induced by BCG.

To assess the performance of the QFT-G test for screening before the initiation and also during anti-TNF treatment.

The QFT-G test (intube method) was used to determine IFNγ production after stimulation with MT-specific antigens (ESAT-6, CFP-10 and TB7.7(p4) proteins) as well as with nonspecific mitogen (PHA). Altogether, 317 patients were investigated (rheumatoid arthritis (RA) = 117, ankylosing spondylitis (AS) = 95, adult juvenile idiopathic arthritis (JIA) = 54, Crohn's disease (CD) = 30, and psoriatic arthritis (PsA) = 21). Fifteen AS patients completed QFT-G assessments prior to anti-TNF and also after 2 and 14 weeks of treatment (before the second and fourth infliximab infusions). One hundred and eight patients had the QFT-G only before they started anti-TNF, six patients before and during therapy, and 203 patients only during the treatment. Tuberculin 2TU was used for the TST and results were read after 48–72 hours.

Out of the total 317 tests, 12 (3.8%) were indeterminate (5 × CD, 4 × RA, 2 × JIA, 1 × AS) (two high spontaneous IFNγ production, 10 low mitogen response). Eleven out of these 12 patients were on combined immunosuppression. Twelve patients (10.5%) (5 × RA, 4 × AS, 2 × CD, 1 × JIA) were positive in the QFT-G in the screening. Four had a negative TST, five a positive, in two the TST was unavailable. From 209 patients investigated during the treatment, 12 were QFT-G-positive. In three of them the initial pretreatment status was known and the QFT-G was negative; however, all three were positive in the TST at 16, 7 and 10 mm. Two of them became QFT-G-positive before the second infliximab infusion, one before the fourth infusion. The remaining nine positive patients had longstanding treatment with different anti-TNF biologicals.

In patients with a positive TST (n = 64), 30 had the QFT-G done before treatment – five had positive results and one an indeterminate result. The remaining 34 TST-positive patients were QFT-G-positive in nine cases and one result was indeterminate. In those who were TST-negative (n = 100), five were positive for the QFT-G. One patient developed TB, TST-negative, QFT-G-positive before therapy.

In patients who were on infliximab and repeatedly investigated, IFNγ production after the nonspecific mitogen stimulation increased: 6.7 ± 3.8 IU/ml, 8.4 ± 2.6 IU/ml, and 8.9 ± 3.4 IU/ml (P = 0.001) before treatment and after 2, and 12 weeks, respectively. No difference was found in IFNγ in longitudinal samples after stimulation with TB antigens.

There is scarce information about QFT-G use before and during anti-TNF treatment. Our first experience shows that the QFT-G may be more sensitive and specific for LTBI detection than the TST, although its real usefulness for screening before anti-TNF treatment needs to be assessed in long-term studies. The QFT-G is valuable once anti-TNF has been initiated as we detected a relatively high number of positive patients who may be in danger of TB development. No decrease of the mitogen-induced capacity in IFNγ production shows that test can be meaningfully used during anti-TNF treatment.