Performance of Interferon-Gamma Release Assays in the Diagnosis of Nontuberculous Mycobacterial Diseases-A Retrospective Survey From 2011 to 2019.

There is an urgent need for precise diagnosis to distinguish nontuberculous mycobacterial (NTM) diseases from pulmonary tuberculosis (PTB) and other respiratory diseases. The aim of this study is to evaluate the diagnostic performance of Interferon-gamma (IFN-γ) release assays (IGRAs), including antigen-specific peripheral blood-based quantitative T cell assay (T-SPOT.TB) and QuantiFERON-TB-Gold-Test (QFT-G), in differentiating NTM infections ( = 1,407) from culture-confirmed PTB ( = 1,828) and other respiratory diseases ( = 2,652). At specie level, 2.56%, 10.73%, and 16.49% of NTM-infected patients were infected by , , and with - complex (MAC), respectively. Valid analyses of T-SPOT.TB (ESAT-6, CFP-10) and QFT-G were available for 37.03% and 85.79% in NTM-infected patients, including zero and 100% (36/36) of infection, 21.85% (33/151) and 92.05% (139/151) of infection, and 17.67% (41/232) and 91.24% (211/232) of MAC infection. Based on means comparisons and further ROC analysis, T-SPOT.TB and QFT-G performed moderate accuracy when discriminating NTM from PTB at modified cut-off values (ESAT-6 < 4 SFCs, CFP-10 < 3 SFCs, and QFT-G < 0.667 IU/ml), with corresponding AUC values of 0.7560, 0.7699, and 0.856. At species level of NTM, QFT-G effectively distinguished between MAC (AUC=0.8778), (AUC=0.8834) or (AUC=0.8783) than T-SPOT.TB. No significant differences in discriminatory power of these three IGRA tools were observed when differentiating NTM and Controls. Our results demonstrated that T-SPOT.TB and QFT-G were both efficient methods for differentiating NTM disease from PTB, and QFT-G possessed sufficient discriminatory power to distinguish infections by different NTM species.