SummaryBackgroundHIV-1 infection is associated with increased risk of tuberculosis and a safe and effective vaccine would assist control measures. We assessed the safety, immunogenicity, and efficacy of a candidate tuberculosis vaccine, modified vaccinia virus Ankara expressing antigen 85A (MVA85A), in adults infected with HIV-1.MethodsWe did a randomised, double-blind, placebo-controlled, phase 2 trial of MVA85A in adults infected with HIV-1, at two clinical sites, in Cape Town, South Africa and Dakar, Senegal. Eligible participants were aged 18–50 years, had no evidence of active tuberculosis, and had baseline CD4 counts greater than 350 cells per μL if they had never received antiretroviral therapy or greater than 300 cells per μL (and with undetectable viral load before randomisation) if they were receiving antiretroviral therapy; participants with latent tuberculosis infection were eligible if they had completed at least 5 months of isoniazid preventive therapy, unless they had completed treatment for tuberculosis disease within 3 years before randomisation. Participants were randomly assigned (1:1) in blocks of four by randomly generated sequence to receive two intradermal injections of either MVA85A or placebo. Randomisation was stratified by antiretroviral therapy status and study site. Participants, nurses, investigators, and laboratory staff were masked to group allocation. The second (booster) injection of MVA85A or placebo was given 6–12 months after the first vaccination. The primary study outcome was safety in all vaccinated participants (the safety analysis population). Safety was assessed throughout the trial as defined in the protocol. Secondary outcomes were immunogenicity and vaccine efficacy against Mycobacterium tuberculosis infection and disease, assessed in the per-protocol population. Immunogenicity was assessed in a subset of participants at day 7 and day 28 after the first and second vaccination, and M tuberculosis infection and disease were assessed at the end of the study. The trial is registered with ClinicalTrials.gov, number NCT01151189.FindingsBetween Aug 4, 2011, and April 24, 2013, 650 participants were enrolled and randomly assigned; 649 were included in the safety analysis (324 in the MVA85A group and 325 in the placebo group) and 645 in the per-protocol analysis (320 and 325). 513 (71%) participants had CD4 counts greater than 300 cells per μL and were receiving antiretroviral therapy; 136 (21%) had CD4 counts above 350 cells per μL and had never received antiretroviral therapy. 277 (43%) had received isoniazid prophylaxis before enrolment. Solicited adverse events were more frequent in participants who received MVA85A (288 [89%]) than in those given placebo (235 [72%]). 34 serious adverse events were reported, 17 (5%) in each group. MVA85A induced a significant increase in antigen 85A-specific T-cell response, which peaked 7 days after both vaccinations and was primarily monofunctional. The number of participants with negative QuantiFERON-TB Gold In-Tube findings at baseline ...