HIV-1 infection impairs tuberculosis (TB) specific immune responses affecting the diagnosis of latent TB. We aimed to (1) determine the proportion of HIV-1-infected adults with a negative QuantiFERON®-TB Gold in-tube (QFT-GIT) and Tuberculin skin testing (TST) that convert to QFT-GIT positive following TST, and (2) evaluate the relationship between conversion and baseline QFT-GIT results.
HIV-1 infected adults being screened for a TB vaccine study in South Africa underwent QFT-GIT followed by TST. As per protocol, QFT-GIT was repeated if randomization was delayed allowing for evaluation of TST boosting in a proportion of participants.
Of the 22 HIV-1 infected, TST and QFT-GIT negative adults (median CD4 477/mm(3) IQR 439-621) who had QFT-GIT repeated after median 62 days (IQR 49-70), 40.9 % (95 % CI 18.6-63.2 %) converted. Converters had a significantly greater increase in the background subtracted TB antigen response (TBAg-Nil – all units IU/mL) following TST, 0.82 (IQR 0.39-1.28) vs 0.03 (IQR -0.05-0.06), p = 0.0001. Those who converted also had a significantly higher baseline TBAg-Nil 0.21(IQR 0.17-0.26) vs 0.02(IQR 0.01-0.07), p = 0.002. Converters did not differ with regard to CD4 count or ART status. ROC analysis showed a baseline cut off of 0.15 correctly classified 86.4 % of converters with 88.9 % sensitivity.
Our findings support the possibility that there are 2 distinct groups in an HIV-1 infected population with negative QFT-GIT and TST; a true negative group and a group showing evidence of a weak Mtb specific immune response that boosts significantly following TST resulting in conversion of the test result that may represent false negatives. Further evaluation of whether a lower cut off may improve sensitivity of QFT-GIT in this population is warranted.