Lack of efficacy of isoniazid (INH) prophylaxis and PK evaluation in South African infants

Polly Clayden, HIV i-Base

A South African study looked at isoniazid (INH) prophylaxis in young infants and found no increase in TB free survival. [1]

PACTG 1041 was a phase II/III double blind, randomised, placebo-controlled study of primary INH prophylaxis for prevention TB disease and latent infection infants with perinatal HIV-exposure.

In this study, HIV-positive, BCG vaccinated infants of 3-4 months of age were randomised to daily INH (10-20mg/kg/day) or placebo for 96 weeks. The infants also received cotrimoxazole, and ART where indicated, in accordance with WHO guidelines.

The primary objective of the study was to investigate whether INH increases TB disease-free survival in young infants.

Endpoints were TB disease and mortality at 96 weeks.

HIV-positive children (n=452, 226 per arm) with median age 96 days were enrolled between December 2004 and March 2008.

The children’s baseline median CD4% was 27% (range: 6-58%); 91% were CDC clinical category N/A, their median viral load was 666,500 copies/mL and 28% were receiving ART. At time of this scheduled interim analysis 66% of children were receiving ART.

The investigators found, at a median of 36 weeks follow up, 39 (17.3%) and 32 (14.2%) children in the INH and placebo groups, respectively, had TB or died, p=0.34. There were 24 (10.6%) and 22 (8.4%) (p=0.69) cases of TB and 15 (6.6%) and 10 (4.4%) non-TB related deaths in the INH and placebo groups, respectively. They reported no significant difference in rates of adverse event rates between the two groups.

In this study the overall cumulative incidence of TB by 96 weeks was high (22.2%; 95%CI: 15.7, 31.0).

The investigators wrote: “INH prophylaxis did not improve TB-disease free survival in HIV-positive African children with access to ART, indicating the need for alternative strategies to reduce the high public-health burden of childhood TB.”

As the appropriate INH infant dose is unknown, PACTG 1041 also investigated INH PK, and determined N-acetyltransferase-2 (NAT2) genotype to evaluate if PG explains INH PK [2].

The PK study target enrollment is 336 infants. Half of the infants were sampled at weeks 0 and 84 at 2 and 4 hours post dose, and the remaining children at weeks 12 and 84 at 1 and 3 hours post dose. INH was quantified in plasma (HPLC). NAT2 genotype was determined using RFLP and phenotypes assigned as slow (S), intermediate (I), and fast (F) acetylators.

This study used a 1-compartment model with first-order absorption and elimination (NONMEM v.VI). Covariates, including NAT2 phenotype, age, weight, sex, and HIV status, were evaluated using stepwise forward inclusion (p=0.05) and backward elimination (p=0.01).

The investigators modeled 306 INH concentrations from 131 infants. The infants had a median age of 171 days (range 91-717 days) at sampling; 53 were HIV-positive; 65 were girls; NAT2 phenotype, 32 S, 46 I, 30 F. Mean (SD) INH dose, 14 (3) mg/kg/d. Mean (SD) INH concentrations at 1, 2, 3, and 4 hours post dose were 12.0 (4.7), 8.3 (3.8), 6.3 (3.0), and 4.4 (3.0) mg/L, respectively.

They found the infants’ weight and NAT2 phenotype but not HIV status explained most of the interpatient variability in INH oral clearance (CL/F). Typical CL/F at weeks 0 and 12 for F phenotype were 3.3 and 3.9 L/hr and were 1.4 and 1.7 L/hr for S.

They wrote: “INH PK at a dose of 10-20 mg/kg/d in these infants are similar to published data in older (median 3.8 years) children receiving 10 mg/kg/d. The comparability of PK supports continued evaluation of this dose, which is at least twice that recommended by WHO.”


The full results of this trial are not yet out (but are unlikely to change). The investigators are studying the data to explain why no benefit was found for pre-exposure prophylaxis despite a high rate of incident TB.


  1. Madhi SA, Nachman S, Violari A et al. Lack of efficacy of primary isoniazid (INH) prophylaxis in increasing tuberculosis (TB) free survival in HIV-infected (HIV+) South African children. 48th ICAAC, 25-28 October 2008. Washington. Abstract G2-1346a.
  2. Kiser J, Zhu R, Nachman S et al. Pharmacokinetics (PK) and Genetics (PG) of Isoniazid (INH) in South African HIV-Exposed Infants-PACTG 1041. 48th ICAAC, 25-28 October 2008. Washington. Abstract A-1826.

Links to other websites are current at date of posting but not maintained.