Superior Sensitivity of Ex Vivo IFN-γ Release Assays as Compared to Skin Testing in Immunocompromised Patients

Comparative assessment of the tuberculin skin testing (TST) and commercial IFN-γ release-assays (IGRAs) is hampered by the use of different antigens (tuberculin PPD in TST vs. ESAT-6/CFP-10 in IGRAs). Thus, PPD was used as a common stimulus to compare performance of the TST and three IGRAs in 72 controls, 101 hemodialysis patients and 100 renal transplant recipients. Results of the TST were compared with PPD-induced IFN-γ induction in vitro detected by ELISPOT, ELISA or a flow-cytometric FACS assay. Percentages of positive tests were significantly lower in TST (9.2%) compared to ELISA (55.3%), ELISPOT (45.3%) and FACS (44.9%, p < 0.0001). Agreement between TST and IGRAs was highest for controls (κ = 0.19-0.32) and poor in immunocompromised patients (κ = 0 for transplant patients, κ = 0.06-0.13 for hemodialysis patients). Discrepant results were largely TST negative and IGRA positive. Among IGRAs, agreement was highest between ELISPOT and FACS (κ = 0.61). Unlike TST, all IGRAs were associated with variables of mycobacterial exposure. Among IGRAs, the FACS assay was least affected by the level of immunosuppression. In conclusion, both the percentage of positive results and between-test-agreement were higher with IGRAs as compared to TST. This indicates superiority of IGRAs in detecting a PPD-specific immune response which may also apply for immunity toward Mycobacterium tuberculosis-specific antigens.