859
Performance Characteristics of a Whole Blood InterferonG Release Assay in HIV Patients Commencing ART in a High TB Incidence Setting
Julian Elliott*1, S Sarun2, S Chin3, S Chel2, S Huffam1,2, S Huffam1,2, V Saphonn2,3, V Saphonn2,3, J Kaldor1, M French4,5, M French4,5, D Cooper1, and C Mean2
1Natl Ctr for HIV Epidemiology and Clin Res, Univ of New South Wales, Australia; 2Natl Ctr for HIV/AIDS, Dermatology and STDs, Ministry of Hlth, Phnom Penh, Cambodia; 3Natl Inst of Publ Hlth, Ministry of Hlth, Phnom Penh, Cambodia; 4Univ of Western Australia, Perth; and 5PathWest Immunology, Perth, Australia
Background: Whole blood interferon gamma (IFN-γ) release assays are
used for the diagnosis of latent tuberculosis (TB) infection and may have
improved performance, compared to tuberculin skin testing, in people with HIV.
Some data, however, suggest low reproducibility in this group and the effect of
ART is not known.
Methods: We prospectively investigated IFN-γ
responses in people with HIV commencing ART at an ambulatory clinic in Cambodia. A
QuantiFERON-TB Gold In Tube (QFTGIT) assay was
performed prior to ART, at months 1, 3, and 6 of treatment, and at time of
suspected mycobacterial disease. We compared QFTGIT responses in people
receiving treatment for TB, with previous TB, and without TB at these time
points pre- and post-ART.
Results: The study population comprised 186 adults, of
whom 85 (46%) were female. Median age was 33 years and median pre-ART CD4
T-cell count was 63/mm3. Prior to enrolment or during follow-up, 54
(29%) patients had ≥1 episodes of TB treatment Overall, 16% (78 of 481)
of QFTGIT assays were positive. A positive result was associated with higher
CD4 T-cell count (p <0.001), use
of ART (19% vs 11% pre-ART; p = 0.02)
and TB status (p <0.001). Among
the 37 patients tested during TB treatment, 14 (38%) tested positive, including
5 patients (14%) that converted from negative to positive following initiation
of ART. We tested during the first month of TB treatment, 13 patients with a
median CD4 T-cell count of 57/mm3 of whom 6 (46%) tested positive.
In comparison, the proportion of patients with no TB history testing positive
was 18 of 132 (14%). Indeterminate results were seen in 16% and 20% of pre- and
post-ART assays respectively and were associated with lower CD4 T-cell counts (p <0.001). Among 113 patients with at
least 3 QFTGIT results and 3 months post-ART follow up, QFTGIT results remained
unchanged in 99 (88%), converted in the context of active TB in 6 (5%) and
converted independent of active TB in 8 (7%). Median change in response between
pre-ART and month 3 of ART was +0.01 IU/mL (IQR –0.03 to +0.08) and was not
associated with TB status or CD4 T-cell count.
Conclusions: Positive QFTGIT assays were more common in
people with higher CD4 T-cell counts or receiving ART or treatment for TB.
Indeterminate results were relatively common in people with low CD4 counts
prior to and during early ART. QFTGIT responses remained stable following
introduction of ART in the majority of patients.
|
