Plasma Pharmacokinetics of High-Dose Oral versus Intravenous Rifampicin in Patients with Tuberculous Meningitis: a Randomized Controlled Trial

Abstract

Higher doses of intravenous rifampicin may improve outcomes in tuberculous meningitis but are impractical in high-burden settings. We hypothesized that plasma rifampicin exposures would be similar between oral dosing of 35 mg/kg of body weight and intravenous dosing of 20 mg/kg, which has been proposed for efficacy trials in tuberculous meningitis. We performed a randomized parallel-group pharmacokinetic study nested within a clinical trial of intensified antimicrobial therapy for tuberculous meningitis. HIV-positive participants with tuberculous meningitis were recruited from South African hospitals and randomized to one of three rifampicin dosing groups: standard (oral 10 mg/kg), high dose (oral 35 mg/kg), and intravenous (20 mg/kg). Intensive pharmacokinetic sampling was done on day 3. Data were described using noncompartmental analysis, and exposures were compared by geometric mean ratios (GMRs). Forty-six participants underwent pharmacokinetic sampling (standard dose, n = 17; high-dose oral, n = 15; intravenous, n = 14). The median CD4 count was 130 cells/mm³ (interquartile range [IQR], 66 to 253 cells/mm³). The rifampicin geometric mean area under the concentration-time curve from 0 to 24 h (AUC_0-24) values were 42.9 μg · h/ml (95% confidence interval [CI], 24.5 to 75.0 μg · h/ml) for the standard dose, 295.2 μg · h/ml (95% CI, 189.9 to 458.8 μg · h/ml) for the high oral dose, and 206.5 μg · h/ml (95% CI, 154.6 to 275.8 μg · h/ml) for intravenous administration. The rifampicin AUC_0-24 GMR was 1.44 (90% CI, 0.84 to 2.21) and the maximal concentration of drug in serum (C_max) GMR was 0.89 (90% CI, 0.63 to 1.23) for high-dose oral administration with respect to intravenous dosing. The plasma rifampicin AUC_0-24 was higher after an oral 35-mg/kg dose than with intravenous administration at a 20-mg/kg dose over the first few days of tuberculosis (TB) treatment. The findings support oral rifampicin dosing in future tuberculous meningitis trials.

Trial registration: ClinicalTrials.gov NCT03927313.

Keywords: human immunodeficiency virus; pharmacokinetics; rifampicin; tuberculous meningitis.

FIG 1

Trial consort flow diagram. Arm 1, standard TB therapy; arm 2, high-dose rifampicin plus linezolid; arm 3, high-dose rifampicin plus linezolid and aspirin. IPK, intensive PK; AUC, area under the concentration-time curve up to 24 h. Adequate PK profiles are those with at least two observations in the elimination phase.

FIG 2

Individual concentration-time profiles. PK profiles for all participants by rifampicin dose allocation are shown. Gray lines indicate individual profiles, and colored dashed lines indicate geometric means. IV, intravenous.

FIG 3

Comparison of exposures across dosing groups. Open circles are individual values for the AUC_0–24 (A) and C_max (B), boxes indicate medians and interquartile ranges, and whiskers indicate the upper adjacent values (1.5× IQR).

FIG 4

Bioequivalence plot. Point estimates of geometric mean ratios (GMRs) for the AUC_0–24 and C_max, with 90% confidence intervals, are shown, with vertical lines indicating bioequivalence margins. The reference measure is intravenous administration (û_oral/û_i.v.); therefore, a value of >1 favors oral dosing.

FIG 5

Probability density distributions for efficacy target attainment of rifampicin with different dosing strategies. The solid vertical line on the x axis represents the putative efficacy target AUC_0–24 of 203 μg · h/ml.

FIG 6

Simulated exposures across LASER-TBM weight bands for 35-mg/kg dosing, with observed exposures superimposed. Boxes indicate medians and interquartile ranges, and whiskers indicate ranges for simulated exposures derived from external cohorts, as described in the text. Red circles indicate observed exposures from the LASER-TBM cohort.