Prospective randomized crossover study on the performance of QuantiFERON®-TB Gold In-Tube test for screening and monitoring of latent tuberculosis infection across patients undergoing anti-TNF treatment

Background: Screening for latent tuberculosis infection (LTBI) prior to treatment with anti-TNF therapy, as well as a watchful monitoring during the treatment, is strongly recommended. Objectives: To investigate the performance of QuantiFERON®-TB Gold In-Tube (QFT-GIT) and its agreement with tuberculin skin test (TST) in patientsawaiting biologic agents, and to evaluate the usefulness of serial QFT-GIT testing in identifying patients with reactivation of LTBI or those exposed to Mycobacterium tuberculosis, thus analyzing possible conversions or reversions following the commencement of TNF blockers. Methods: We conducted a prospective, randomized, crossover study on consecutive patients with chronic inflammatory rheumatic diseases designated to initiate treatment with anti-TNF agents. Patients were randomized to undergo, on the same day, QFT-GIT and TST by means of a randomization list. In those with evidence of TB infection, after ruling out active TB, LTBI was diagnosed, and a 9-month course isoniazid (INH) prophylaxis was prescribed before the initiation of TNF antagonists. QFT-GIT was repeated after 3 and 6 months since TNF antagonists onset. Results: Among patients enrolled (n=119; M/F=37/82; median age=47 years, standard deviation=15.8), 24 (20.2%) had at least one risk factor for LTBI. A total 96 patients (81%) were under treatment with immunosuppressive drugs at the time of execution of TB testing. 21 (17.6%) were considered to be affected by LTBI due to one abnormal screening tool as a minimum.At baseline, 5 patients (4%) displayed positive, 94 (79%) negative, and 20 (17%) indeterminate QFT-GIT results, while the TST was abnormal in 14 (11.7%). The level of agreement between the 2 tests was 85.8% (k: 0.16). After 6 months, follow-up was complete in 12 patients with LTBI and in 73 without TB. In the first group, the QFT-GIT assay was positive in 2 (16%) patients at month 3 and in 3 (25%) at month 6. No assay was deemed indeterminate. At month 6 we found a conversion of QFT-GIT result in 3 patients and a reversion in 1 patient. In those without evidence of LTBI, the QFT-GIT assay displayed a conversion in 5 (6.8%) patients at month 3. After 6 months, the QFT-GIT assay remained positive in 3 of 5 converters, while the other 2 subjects turned QFT-GIT negative. Overall view of the data showed that, after the beginning of anti-TNF treatment, the QFT-GIT results changed of 28% at month 3 and of 21% at month 6. The greatest statistically significant change was observed in patients with indeterminate results that became negative (15%; p<0.02). There was also a significant increase in IFN-g concentrations in patients who converted their response to positive (p<0.01). No active TB cases were detected. Conclusions: The poor correlation between QFT-GIT and TST test at baseline suggests that the two tests identify different populations of patients and it may be prudent to use both in subjects scheduled for anti-TNF treatment. The application of QFT-GIT in serial testing calls for a careful evaluation of conversions and reversions, and highlights the need to define the within-subject variability of the responses and which variables may predict the conversion even in patients with a low TB risk.