The current TB treatment as recommended by World Health Organization (WHO) although capable of achieving 85% cure rates, has limitations, in particular drug interactions, toxicities, and the long treatment duration which increases the possibility of nonadherence.

Sub-therapeutic isoniazid concentrations were demonstrated in several studies, including our previous work, carried out among patients with tuberculosis receiving the standard dose (5mg/kg) of isoniazid. The investigators found 78% of patients with HIV had isoniazid concentrations below the recommended threshold. Malabsorption, drug-drug interactions, poor adherence due to high pill burden may contribute to this. Pharmacogenetic variation may compound these factors; isoniazid displays inter-individual variation in serum concentrations and clearance due to differences in individual acetylator status.

While patients who metabolize isoniazid slowly (slow acetylators) are at a higher risk of high drug concentrations and toxicities, fast acetylators are more likely to have sub-therapeutic isoniazid concentrations.

In other studies, insufficient exposure with isoniazid, one of the cornerstone drugs for TB treatment, has been associated with delayed sputum clearance, development of drug resistance, and treatment failure.

Isoniazid is metabolized by the enzyme N-acetyl transferase, which in turn is controlled by the N-acetyl transferase-2 (NAT-2) gene. Polymorphisms in this gene are responsible for the N-acetylation phenotypes, with the distribution of NAT-2 fast, intermediate, and slow acetylators being highly variable especially among African populations.

Given that NAT2 acetylator status explains most of the variability in INH exposures, knowledge of NAT2 status may be a simpler way to select the right dose for individual patients. The investigators will therefore provide higher doses to fast acetylators and compare the isoniazid pharmacokinetics in these patients to slow acetylators who receive the standard dose, who are more likely to already be achieving target concentrations.

N-acetyl transferase genotyping will be performed on HIV infected patients diagnosed with T B. Slow acetylators will be initiated on the standard dose of isoniazid (5mg/kg) while fast and intermediate acetylators will be initiated on 10mg/kg of isoniazid. All other TB drugs will be given at their standard doses.

Monitoring for toxicities will be performed every two weeks (including ALT and screening for peripheral neuropathy) and isoniazid concentrations will be measured four weeks after initiating treatment at Ohr, 30mins, 1 hr, Thr and 4hr following observed drug intake.

Patients will be continued on standard dose isoniazid after completing the first 8 weeks of treatment (intensive phase).

Sputum cultures for mycobacteria will be performed at baseline, week 2, 4 and week 8. TB treatment outcome will be assessed after 6 months of treatment.

Non-linear mixed effects modelling will be used to model PK-PD data taking into account clinical and demographic factors like age, sex and BMI. The investigators will develop a model to establish the population parameters for isoniazid (for example clearance, absorption rate constant and volume of distribution) and the variability around these primary PK parameters. The investigators will then use the model to derive secondary PK parameters, namely area under the concentration-time curve (AUC) and maximum concentrations (Cmax) of isoniazid for each participant. The investigators will compare the PK parameters of isoniazid in fast/intermediate acetylators taking 10mg/kg and slow acetylators taking 5mg/kg of isoniazid.

In addition to the main objectives of the study, the investigators will also compare the PK in patients in the PG guided isoniazid dosing group to those in the historical cohort while matching for NAT2 status, age and sex. In the historical cohort, NAT-2 acetlyator genotyping was performed but patients received standard dosing regardless of NAT-2 acetylator status. In this trial, the investigators will use pharmacogenetic guided therapy for all participants. The investigators will therefore be able to compare the pharmacokinetic data above, safety and efficacy in patients who received pharmacogenetic therapy and those who did not.

The investigators will use PK-PD models to describe the relationship between concentrations and pharmacodynamic data (including toxicities and sputum conversion at week 8) in patients on the different doses. In addition to this, the investigators will also make this same comparison for patients in this study and those who did not receive PG guided therapy from the historical cohort. The investigators will also compare the number of grade 3-5 adverse events and time to sputum culture conversion in patients in patients who did and did not receive PG guided therapy while matching for NAT2 status, age and sex

Inclusion Criteria:

• Evidence of a personally signed and dated informed consent document indicating that the subject (or a legal representative) has been informed of all pertinent aspects of the study.
• Subjects who are willing and able to comply with scheduled visits, treatment plan, laboratory tests, and other study procedures.
• Age of ≥18 years
• Bacteriologically confirmed pulmonary TB (determined by Xpert, culture, or microscopy)
• Confirmed HIV-1 infection.
• On TB treatment for ≤ 7 days at the time of enrolment (Within this time, the patient is still expected to have mycobacteria present in sputum and will provide enough time to conduct screening procedures)

Exclusion Criteria:

• TB infection of any organ/systems requiring TB treatment longer than 6 months
• Pregnancy
• Decompensated liver disease and/or aminotransferases >2.5 x ULN

Individual patient data (IPD) will made available to other researchers for further analysis or met-analysis upon reasonable request to the investigators.

6 months after publication of study results.

A direct request shall be made to the investigators .