Laura Dickinson1, Alan Winston2, Marta Boffito3, Saye Khoo1, David Back1, Marco Siccardi1. 1. Department of Molecular & Clinical Pharmacology, University of Liverpool, Liverpool, UK. 2. Faculty of Medicine, Imperial College, London, UK; Department of HIV & Genitourinary Medicine, Imperial College Healthcare NHS Trust St Mary's Hospital, London, UK. 3. Faculty of Medicine, Imperial College, London, UK; St Stephen's Centre, Chelsea & Westminster Foundation Trust, London, UK.
Abstract
INTRODUCTION: Treatment of HIV/TB co-infection is challenging due to high drug-drug interaction potential between antiretrovirals and rifamycins, such as rifampicin (RIF). The PK interaction between darunavir/ritonavir (DRV/RTV) and RIF has not been studied. Utilizing other protease inhibitor data, population PK modelling and simulation was applied to assess the impact of RIF on DRV/RTV PK and generate alternative dosing strategies to aid future clinical trial design. MATERIALS AND METHODS: A previously developed model describing DRV/RTV PK including data from three studies in HIV patients was used [n=51, 7 female, DRV/RTV 800/100 mg (n=32) or 900/100 mg once daily (qd; n=19) (1). The PK interaction between DRV/RTV and RIF was assumed to mimic that observed in HIV-infected, TB negative patients receiving lopinavir (LPV)/RTV (n=21) (2). Simulations of DRV/RTV 800/100 mg qd (n=1000) were performed (-RIF). The model was adapted to increase the typical value of apparent oral clearance (CL/F) by 71% and 36% and decrease relative bioavailability (F) by 20% and 45% for DRV and RTV, respectively (2); 1000 simulations were generated (+RIF). Dose adjustments of DRV/RTV 1200/200 mg qd, 800/100 mg and 1200/150 mg twice daily (bid) were simulated to overcome the interaction. DRV trough (Ctrough) for each dosing scenario was compared to the reference (-RIF) by GMR (90% CI). RESULTS: DRV and RTV were described by a 1 and 2-compartment model, respectively. A maximum effect model, with RTV inhibiting DRV CL/F, best described the relationship between the drugs. Compared to the reference (-RIF), simulated DRV Ctrough was 70%, 46% and 20% lower for 800/100 mg qd, 1200/200 mg qd and 800/100 mg bid all +RIF, respectively. Ctrough was 38% higher with 1200/150 mg bid +RIF (Table 1). CONCLUSIONS: Modelling and simulation was used to investigate the theoretical impact of RIF on DRV/RTV PK. Based on simulations, 800/100 mg and 1200/150 mg both bid could largely overcome the impact of the interaction. However, the risk of increased RTV-related side effects and higher pill burden should be considered. In vitro work is ongoing to develop a physiologically based model characterizing the interaction and informing simulations.
INTRODUCTION: Treatment of HIV/TB co-infection is challenging due to high drug-drug interaction potential between antiretrovirals and rifamycins, such as rifampicin (RIF). The PK interaction between darunavir/ritonavir (DRV/RTV) and RIF has not been studied. Utilizing other protease inhibitor data, population PK modelling and simulation was applied to assess the impact of RIF on DRV/RTV PK and generate alternative dosing strategies to aid future clinical trial design. MATERIALS AND METHODS: A previously developed model describing DRV/RTV PK including data from three studies in HIVpatients was used [n=51, 7 female, DRV/RTV 800/100 mg (n=32) or 900/100 mg once daily (qd; n=19) (1). The PK interaction between DRV/RTV and RIF was assumed to mimic that observed in HIV-infected, TB negative patients receiving lopinavir (LPV)/RTV (n=21) (2). Simulations of DRV/RTV 800/100 mg qd (n=1000) were performed (-RIF). The model was adapted to increase the typical value of apparent oral clearance (CL/F) by 71% and 36% and decrease relative bioavailability (F) by 20% and 45% for DRV and RTV, respectively (2); 1000 simulations were generated (+RIF). Dose adjustments of DRV/RTV 1200/200 mg qd, 800/100 mg and 1200/150 mg twice daily (bid) were simulated to overcome the interaction. DRV trough (Ctrough) for each dosing scenario was compared to the reference (-RIF) by GMR (90% CI). RESULTS:DRV and RTV were described by a 1 and 2-compartment model, respectively. A maximum effect model, with RTV inhibiting DRV CL/F, best described the relationship between the drugs. Compared to the reference (-RIF), simulated DRV Ctrough was 70%, 46% and 20% lower for 800/100 mg qd, 1200/200 mg qd and 800/100 mg bid all +RIF, respectively. Ctrough was 38% higher with 1200/150 mg bid +RIF (Table 1). CONCLUSIONS: Modelling and simulation was used to investigate the theoretical impact of RIF on DRV/RTV PK. Based on simulations, 800/100 mg and 1200/150 mg both bid could largely overcome the impact of the interaction. However, the risk of increased RTV-related side effects and higher pill burden should be considered. In vitro work is ongoing to develop a physiologically based model characterizing the interaction and informing simulations.
Summary of model simulated DRV Ctrough concentrations in the absence and presence of RIF and following dose adjustment in combination with RIF. The changes in simulated DRV Ctrough are presented as GMR (90% CI)
Table 1
Summary of model simulated DRV Ctrough concentrations in the absence and presence of RIF and following dose adjustment in combination with RIF. The changes in simulated DRV Ctrough are presented as GMR (90% CI)
Authors: Chao Zhang; Paolo Denti; Eric Decloedt; Gary Maartens; Mats O Karlsson; Ulrika S H Simonsson; Helen McIlleron Journal: Br J Clin Pharmacol Date: 2012-05 Impact factor: 4.335