Effect of isoniazid preventive therapy on immune responses to mycobacterium tuberculosis: an open label randomised, controlled, exploratory study

Irene Andia Biraro, Moses Egesa, Simon Kimuda, Steven G. Smith, Frederic Toulza, Jonathan Levin, Moses Joloba, Achilles Katamba, Stephen Cose, Hazel M. Dockrell, Alison M Elliott

With the renewed emphasis to implement isoniazid preventive therapy (IPT) in Sub-Saharan Africa, we investigated the effect of IPT on immunological profiles among household contacts with latent tuberculosis. Household contacts of confirmed tuberculosis patients were tested for latent tuberculosis using the QuantiFERON®-TB Gold In-Tube (QFN) assay and tuberculin skin test (TST). HIV negative contacts aged above 5 years, positive to both QFN and TST, were randomly assigned to IPT and monthly visits or monthly visits only. QFN culture supernatants from enrolment and six months' follow-up were analysed for M.tb-specific Th1, Th2, Th17, and regulatory cytokines by Luminex assay, and for M.tb-specific IgG antibody concentrations by ELISA. Effects of IPT were assessed as the net cytokine and antibody production at the end of six months. Sixteen percent of contacts investigated (47/291) were randomised to IPT (n = 24) or no IPT (n = 23). After adjusting for baseline cytokine or antibody responses, and for presence of a BCG scar, IPT (compared to no IPT) resulted in a relative decline in M.tb-specific production of IFN gamma (adjusted mean difference at the end of six months (bootstrap 95 % confidence interval (CI), p-value) -1488.6 pg/ml ((-2682.5, -294.8), p = 0.01), and IL- 2 (-213.1 pg/ml (-419.2, -7.0), p = 0.04). A similar decline was found in anti-CFP-10 antibody levels (adjusted geometric mean ratio (bootstrap 95 % CI), p-value) 0.58 ((0.35, 0.98), p = 0.04). We found no effect on M.tb-specific Th2 or regulatory or Th17 cytokine responses, or on antibody concentrations to PPD and ESAT-6. IPT led to a decrease in Th1 cytokine production, and also in the anti CFP-10 antibody concentration. This could be secondary to a reduction in mycobacterial burden or as a possible direct effect of isoniazid induced T cell apoptosis, and may have implications for protective immunity following IPT in tuberculosis-endemic countries. ISRCTN registry, ISRCTN15705625 . Registered on 30(th) September 2015.