Determination of Adequate Tuberculosis Regimen in Patients Hospitalized With HIV-associated Severe Immune Suppression

Settings: 5 African (Cameroon, Guinea, Uganda, Zambia, Mozambique) and 1 South-East Asian (Cambodia) countries. Sample size : 1330 patients (665 in each arm). Follow-up : 48 weeks after entry in the trial (TB treatment initiation). All participants will initiate antiretroviral therapy (ART) 2 weeks after starting TB treatment. In each country, the chosen ART regimen will be the same in both arms. According to the first-line regimen recommended in each country, the ART combination will be TDF/3TC/EFV 600 mg, or TDF/3TC + double-dose DTG. The primary objective is to estimate the impact of an intensified initial phase of TB treatment on mortality at 48 weeks among HIV-infected adults and adolescents hospitalized for TB with CD4 ≤ 100 cells/μL in comparison with standard TB treatment. The secondary objectives are to estimate the impact of an intensified initial phase of TB treatment, in comparison with the standard TB regimen, on: * Mortality at weeks 8 and 24 * Adverse events, including * All grade 3 and 4 events * Selected grade 2 events of interest * Drug-related adverse events * AIDS-defining illnesses * Paradoxical TB-associated immune reconstitution inflammatory syndrome (IRIS) * TB treatment success * TB recurrence * ART response in terms of virological success and immunological response * Adherence to TB treatment and ART * Peak plasma concentrations of rifampicin and isoniazid (and its N-acetyl-metabolite) at day 3, day 7 and week 2 * Plasma concentrations of efavirenz and dolutegravir at week 4 (i.e. 2 weeks after the onset of ART). A pharmacokinetic sub-study of rifampicin and isoniazid will be carried out in 72 voluntary patients (6 patients/arm/country) at the second week of the main study.