Lopinavir–ritonavir super-boosting in young HIV-infected children on rifampicin-based tuberculosis therapy compared with lopinavir–ritonavir without rifampicin: a pharmacokinetic modelling and clinical study

Abstract

Rifampicin reduces lopinavir concentrations in HIV and tuberculosis co-treated patients. We hypothesised that adding ritonavir to co-formulated lopinavir-ritonavir (4:1) to achieve a one-to-one ratio would overcome this drug-drug interaction in young children. We did a prospective, open-label, one-group, one-sequence study at five sites in three South African provinces. We included HIV-infected children with tuberculosis, a bodyweight of 3-15 kg, and a post-conceptional age of more than 42 weeks. Children received the standard four-to-one ratio of lopinavir-ritonavir in the absence of rifampicin-based anti-tuberculosis treatment, whereas super-boosting of lopinavir-ritonavir with additional ritonavir was given orally twice a day to achieve a one-to-one ratio during rifampicin treatment. The primary outcome was the comparison of the proportion of children with predicted lopinavir morning minimum concentrations (Cmin) of more than 1·0 mg/L during super-boosting with the proportion of more than 1·0 mg/L during standard lopinavir-ritonavir treatment without rifampicin. Lopinavir concentrations were determined before and at 1, 2, 4, 6, and 10 h after the morning dose during the second and the last month of tuberculosis co-treatment, and 4-6 weeks after stopping rifampicin. A non-linear mixed-effects model was implemented to interpret the data and Monte Carlo simulations were used to compare the percentage of lopinavir with morning Cmin values of less than 1·0 mg/L for the two dosing schemes. A non-inferiority margin of 10% was used. This study is registered with ClinicalTrials.gov, number NCT02348177. Between Jan 30, 2013, and Nov 9, 2015, 96 children with a median age of 18·2 months (IQR 9·6-26·8) were enrolled. Of these 96 children, 80 (83%) completed the first three pharmacokinetic evaluations. Tuberculosis therapy was started before antiretrovirals in 70 (73%) children. The model-predicted percentage of morning Cmin of less than 1·0 mg/L after tuberculosis treatment without super-boosting was 8·8% (95% CI 0·6-19·8) versus 7·6% (0·4-16·2) during super-boosting and tuberculosis treatment. The difference of -1·1% (95% CI -6·9 to 3·2), at a non-inferiority margin of 10%, confirmed the non-inferiority of lopinavir trough concentrations during rifampicin co-treatment. 19 serious adverse events were reported in 12 participants. Three deaths and a temporary treatment interruption due to jaundice were unrelated to study treatment. Lopinavir exposure with ritonavir super-boosting in a one-to-one ratio during rifampicin-based tuberculosis treatment was non-inferior to the exposure with lopinavir-ritonavir without rifampicin. Safe and effective, field application of super-boosting is limited by poor acceptability. Access to better adapted solid formulations will most likely facilitate public health implementation of this strategy. DNDi, French Development Agency, UBS Optimus Foundation, and Unitaid.