P85 Use of PBPK modeling for drug-drug interactions prediction of RIFAPENTINE on other drugs by induction of CYP3A4/5 and CYP2B6

Olivier Nicolas , sanofi, Montpellier, France
Helene Vermet , sanofi, Montpellier, France
Brigitte Demers , sanofi, Gentilly, France
Xavier Boulenc , sanofi, Montpellier, France
Eric Sultan , sanofi, Montpellier, France
Fran├žois Donat , sanofi, Montpellier, France
Christine Farenc , sanofi, Montpellier, France
IntroductionRifamycins are sterilizing agents, key for tuberculosis (TB) treatment. Among them rifapentine (RPT), initially approved by FDA for active TB (1998), and later (2014) for the treatment of latent TB infection (LTBI). RPT is given in co-administration with isoniazid (900 mg/900 mg) as a once-weekly regimen and was included in the WHO Guidelines for the management of LTBI. To shorten the duration of active TB treatment,1200 mg/day RPT-based regimens are currently in Phase 3.Rifamycins strongly induce and/or inhibit many drug metabolizing enzymes and transporters in vitro which results in complex Drug-Drug Interaction (DDI) profiles. Modeling of rifamycins DDI may help to improve current practices and guide for the selection of new drug combinations for clinical development.The goal of this work supported by the Predict TB consortium was to qualify a PBPK model using Simcyp® and based on clinical PK data in order to perform DDI predictions at different dosing regimen and in special populations (such as Poor Metabolizer CYP2B6) with the main co-administered drugs of interest. Methodology Simcyp® (V16) was used with healthy volunteer population. The combined RPT and 25-desacetyl-RPT (active metabolite) PBPK model was built using physico-chemical data and in vitro parameters. RPT and 25-desacetyl-RPT are in vitro potent inducers of CYP3A4/3A5, CYP2B6, CYP2C8, CYP2C9 and also inhibitor of CYP2B6, CYP2C9 and CYP2D6 for RPT and inhibitor of CYP2C8 for the 25-desacetyl RPT. Both compounds are also inhibitor of transporters. The poster will focus on the qualification and the use of the model for the prediction of CYP3A4/3A5/2B6-related interactions.The PBPK files for the different substrates came from Simcyp® library (midazolam, efavirenz) or were provided by Simcyp® via CPTR (Critical Path to TB drug Regimen) (bedaquiline) or based on literature (indinavir).The RPT/25-desacetyl-RPT PBPK was built with PK data from multiple ascending dose study and a DDI study with midazolam (CYP3A4/3A5 substrate) at dose for active TB. The PBPK model was then qualified with results from DDI studies with indinavir (CYP3A4 substrate), bedaquilline (CYP3A4 substrate) and efavirenz (CYP2B6/3A4 substrate).DDI predictions were performed for the LTBI dosing regimen for midazolam and bedaquiline, and in CYP2B6 poor metabolizer (PM), for efavirenz.ResultsThe predicted /observed ratio for Cmax and AUC ranged from 0.78 to 1.6 for RPT and from 0.73 to 1.6 for 25-desacetyl-RPT capturing well the CYP3A4/3A5 auto-induction trend during daily dosing regimen.The model predicts well the data from existing DDI studies with midazolam, indinavir and efavirenz at various dosing regimens.Using the RPT/25-desacetyl-RPT PBPK model: DDI predictions at the LTBI dose showed that midazolam AUC was decreased by 86% and bedaquilline AUC was decreased by 31% and DDI predictions at the LTBI dose in CYP2B6 PM showed that efavirenz AUC was decreased up to 15% with less interaction than in CYP2B6 Extensive Metabolizer, even if efavirenz is also CYP3A4 substrate.ConclusionThis combined RPT/25-desacetyl-RPT PBPK model was successfully qualified based on clinical data. RPT remains a potent inducer of CYP3A4 when administered once a week for LTBI treatment. No interaction of RPT on efavirenz is expected in CYP2B6 PM despite CYP3A4 contribution.