In Cambodia the prevalence of both tuberculosis (TB) and Human Immunodeficiency Virus (HIV) infection is high. Data suggest that aggressive management of HIV infection, which includes Anti-Retroviral Therapy (HAART) during treatment of TB, decreases both morbidity and mortality. On the other hand, the use of HAART for patients with TB may cause severe complications due to drug-drug interactions, and occasionally a temporary exacerbation of symptoms. These reactions may be particularly severe when HAART is started soon after the start of TB treatment.

The proposed study aims to determine the optimal time to initiate HAART in previously untreated HIV-infected adult patients with TB and low CD4 cell counts.

In Cambodia the prevalence of both tuberculosis (TB) and Human Immunodeficiency Virus (HIV) infection is high. In 2000, there were approximately 75.000 newly diagnosed TB cases. In 2003, 1.9% of the population was infected with HIV. TB rates in Cambodia are more than double those observed in other developing countries and up to 30 times higher than those currently seen in the USA or Western Europe. It is estimated that over 8% of the newly diagnosed TB cases are co-infected with HIV, of which approximately 85% are severely immunosuppressed (CD4+ cell count < 200 x 106 cells/l).

Mortality rates were found to be 2-4 folds higher in HIV/TB co-infected patients than in TB alone. Data suggest that aggressive management of HIV infection, which includes Highly Active Anti-Retroviral Therapy (HAART) during treatment of TB decreases both morbidity and mortality by suppressing viral replication and improving immune function.

On the other hand, the use of HAART for patients with TB may cause severe complications due to drug-drug interactions, and occasionally a temporary exacerbation of symptoms, signs or radiographic manifestations of TB. Such events or 'paradoxical reactions' that occur among 7 - 36% of HIV/TB co-infected patients treated with HAART may be secondary to immune restitution. These reactions may be particularly severe when HAART is started soon after the start of TB treatment.

Most clinical teams recommend delaying the initiation of HAART to avoid the early side effects of TB treatment and simplify clinical management of the co-infected patient. However others argue that early initiation of HAART in TB patients with CD4 cell counts < 100 x 106 cells/l leads to a marked reduction of viral load despite frequent adverse events.

The proposed study aims to determine the optimal time to initiate HAART (defined as d4T + 3TC + efavirenz) in previously untreated HIV-infected adult patients with TB and low CD4 cell counts. The study is a multicentre prospective, randomized, open-label two-armed trial with no placebo. It is designed as a superiority trial to compare the "early arm" (HAART initiated 2 weeks after TB treatment onset) with the "late arm" (HAART initiated 2 months after TB treatment onset). Efficacy will be assessed by the survival rate in the two arms. Secondary objectives will include evaluation of (1) the safety of an early initiation of HAART in terms of drug interactions or paradoxical reactions, (2) the occurrence of opportunistic infections diagnosed during the follow-up period, (3) patients' adherence to TB treatment and HAART, (4) the rate of hospitalization for any cause during the trial; the measure of (5) the effectiveness of the TB treatment and HAART and (6) the predictive factors for the survival, the response to anti-TB therapy and HAART and the paradoxical reactions.

The total study duration is expected to be 4 years (3 years for enrolment and at least one year of follow-up) in five study sites: (1) Khmero-Soviet Friendship Hospital, Phnom Penh; (2) Calmette Hospital, Phnom Penh; (3) Provincial hospital, Svay Rieng province; and (4) Provincial hospital, Takeo province, (5) Provincial Hospital, Siem Reap.

The study will be carried out in compliance with the protocol and in accordance with the Declaration of Helsinki approved by the World Health Association and with the recommendations of the Good Clinical Practice.

Inclusion Criteria:

• Age 18 or over
• Positive HIV test result
• CD4+ cell count under or equal to 200 cells per ml within 14 days prior the study entry
• Positive AFB on any smear (sputum, lymph node drainage, stool, CSF, pleural fluid)
• Naive to ART
• TB treatment started less than one week prior enrolment
• Negative gonadotrophin pregnancy test (blood) for women of childbearing potential (i.e. not surgically sterile or less than 2 years menopause)
• Agreement from female candidates who are participating in sexual activity that could lead to pregnancy while receiving and for 6 weeks after stopping efavirenz to use two reliable methods of contraception, one of which including condom.

Exclusion Criteria:

• Suspected TB with negative AFB
• Pregnant or breastfeeding women
• Impaired hepatic function (icterus, elevated AST or ALT at least 5 times over the normal value)
• Unable and/or unlikely to comprehend and/or be adherent to the protocol
• Treated for a previous suspected or documented TB other than the ongoing infection which motivates enrolment in this trial