Impact of Tuberculosis Infection on HIV-1 Antibody Response (NWCS501)

Study Objective: Compare longitudinal HIV-1 antibody responses among HIV-1 infected participants that either did or did not develop active TB enrolled in ACTG 5274 and ACTG 5279  

Study Design: Longitudinal antibody responses will be examined among select participants in A5274 who have plasma samples available at the time around and at least 6 - 12 months prior to TB disease diagnosis. All HIV-1/TB participants will be matched to participants that did not develop TB over longitudinal follow up. The non-TB disease comparison group will have relatively similar time interval difference between the 2 longitudinal samples (within 3 – 6 months). The non-TB disease participants will also be matched based on the absolute CD4 count (within 100 cells/mm3) in the earliest sample, and presence or absence of cART over the duration of follow-up.

ACTG 5274 is comprised of adults and adolescents with very advanced HIV CD4<50 cells/mm3 and all initiated ART. In an initial analysis by Dr. Michael Hughes, longitudinal samples are available from 10 confirmed TB cases and 7 probable TB cases with pre-TB plasma samples and also samples close to TB diagnosis.  All cases have possible multiple controls available that are matched for duration of time between pre-TB and follow-up sample, and cART use. In ACTG 5274, cases and controls will also be matched on pre-TB HIV plasma virus level. 
We will compare antibody responses at the follow up time point relative to enrollment. We will also compare antibody responses among the 2 groups (HIV-1/TB disease versus HIV-1 participants that did not develop TB) at the 2 time points using matched-pairs parametric and non-parametric tests. We hypothesize that active TB disease enhances HIV-1 antibody responses.  Thus the cases as compared to controls will have significantly higher change in antibody response over time and higher neutralizing BP score at the time of TB disease diagnosis.  Both groups will have similar HIV-1 antibody responses in the pre-TB sample. These studies will suggest that active TB disease enhances HIV-1 antibody responses. Neutralization assays will be done using standard TZM-bl cells at the highest plasma dilution of 1:50 as documented above. Neutralization responses will be assessed using the BP score against a panel of heterologous HIV-1 envelopes as documented above. Assuming participants that eventually develop TB over longitudinal follow up (HIV-1/TB) as compared to those that did not develop TB have around 2 fold difference in neutralization BP score with the observed standard deviations (Fig. 1A), we will have 80% and 90% power to detect statistically significant difference (p < 0.05) by examining 15 and 18 samples per group respectively.