A T-cell-based enzyme-linked immunospot assay for tuberculosis screening in Chinese patients with rheumatic diseases receiving infliximab therapy

Anti-tumour necrosis factor-α (TNF-α) therapy brought new hopes for treating rheumatic diseases but also increased the risk of infection, including mycobacterium tuberculosis (MTb). Conventional screening tools, such as tuberculin skin test (TST), lack sensitivity or specificity. Recently, T-SPOT.TB has been introduced to detect tuberculosis infection. Reports have proved its superior performance in detecting tuberculosis infection in various patient populations than the TST. To compare the value of a T-cell-based enzyme-linked immunospot assay (ELISPOT) T-SPOT.TB and conventional (TST) in screening and monitoring tuberculosis in patients with rheumatic diseases during infliximab therapy in China. Fifty-eight patients with various rheumatic diseases who received infliximab therapy were enrolled in the trial. Freshly isolated peripheral blood mononuclear cells were stimulated with MTb-specific antigens (ESAT-6 and CFP10), and IFN-γ-producing cells were counted. TST was performed with 1 TU PPD injected intradermally into the volar aspect of forearm. A cutaneous induration with diameter ≥5 mm was considered as positive TST, and an increment ≥5 mm of cutaneous induration was considered as TST conversion. TST and T-SPOT.TB test were carried out at baseline and repeated 12 months after infliximab therapy (if no active TB occurs) or at times when TB occurred. Moreover, all patients were initially evaluated for latent tuberculosis infection (LTBI) with clinical examination and chest radiograph. The McNemar test was used for TST and T-SPOT.TB concordance analysis. Cohen’s kappa coefficient was used to assess strength of the agreement. Among the 58 patients evaluated, 25 (43.1%) had ankylosing spondylitis, 24 (41.4%) had rheumatoid arthritis, 4 (6.9%) had undifferentiated spondyloarthropathy, 3 (5.2%) had psoriatic arthritis and 2 (3.4%) had reactive arthritis. A total of 52 patients (89.7%) had previously received vaccination with Bacille Calmette-Guerin. All of the patients received either single or combination of disease modifying anti-rheumatic drug (DMARDs) therapy, and 16 (27.4%) had previously or presently received glucocorticoid therapy. Before infliximab therapy, 12 patients (20.7%) had positive and 46 (79.3%) had negative TST results, and only 1 (1.7%) had positive T-SPOT.TB. Among 51 patients completing the repeated TST and T-SPOT.TB assay, 7 patients (13.7%) had TST conversion and 4 (7.8%) had positive T-SPOT.TB results. Of 7 patients with TST conversion, 2 patients (28.6%) developed active TB and also had positive T-SPOT.TB results; of 44 patients with no TST conversion, 2 patients (4.5%) had positive T-SPOT.TB and 1 (2.3%) had active TB. If 5 mm was used as the cut-off value of TST, TST and T-SPOT.TB, had an agreement value of 68.6% with a kappa value of 0.166. If 10 mm was used as the cut-off value, the agreement between TST and T-SPOT.TB was 88.2% with a kappa value of 0.338. T-SPOT.TB was more specific than TST in detecting tuberculosis during infliximab therapy in China with high BCG vaccination and high prevalence of TB. It can be used as a reliable tool for TB monitoring during infliximab therapy in Chinese patients with rheumatic diseases. Finally, it is recommended to repeat the TST and T-SPOT.TB periodically during biological treatment.