Drug combinations for colorectal cancer treatment_PBPK

Use of a physiologically based pharmacokinetic model to simulate drug–drug interactions between antineoplastic and antiretroviral drugs Jose´ Molt o 1,2 *, Rajith Rajoli 3 , David Back 3 , Marta Valle 2,4 , Cristina Miranda 1 , Andrew Owen 3 , Bonaventura Clotet 1,2,5,6 and Marco Siccardi 3 1 Lluita contra la Sida Foundation, Hospital Universitari Germans Trias i Pujol, Badalona, Spain; 2 Universitat Aut onoma de Barcelona (UAB), Barcelona, Spain; 3 Department of Molecular and Clinical Pharmacology, University of Liverpool, Liverpool, UK; 4 PKPD Modeling and Simulation, Institut de Recerca de l’Hospital de la Santa Creu i Sant Pau – Institut d’Investigacio Biomedica Sant Pau, Barcelona, Spain; 5 Universitat de Vic (UVic), Vic, Spain; 6 IrsiCaixa Foundation, Hospital Universitari Germans Trias i Pujol, Badalona, Spain *Corresponding author. Lluita contra la Sida Foundation, Hospital Universitari Germans Trias i Pujol, Ctra de Canyet s/n, 08916 Badalona, Spain. Tel: þ34-93 -497-88-87; Fax: þ34-93-465-76-02; E-mail: [email protected] Received 17 June 2016; returned 28 August 2016; revised 7 October 2016; accepted 10 October 2016 Background: Co-administration of antineoplastics with ART is challenging due to potential drug–drug interactions (DDIs). However, trials specifically assessing such DDIs are lacking. Our objective was to simulate DDIs between the antineoplastics erlotinib and gefitinib with key antiretroviral drugs and to predict dose adjustments using a physiologically based pharmacokinetic (PBPK) model. Methods: In vitro data describing chemical properties and pharmacokinetic processes of each drug and their effect on cytochrome P450 isoforms were obtained from the literature. Plasma drug–concentration profiles were simulated in a virtual population of 50 individuals receiving erlotinib or gefitinib alone or with darunavir/ritonavir, efavirenz or etravirine. Simulated pharmacokinetic parameters and the magnitude of DDIs with probe drugs (midazolam, maraviroc) were compared with literature values. Erlotinib and gefitinib pharmacokinetics with and without antiretrovirals were compared and dose-adjustment strategies were evaluated. Results: Simulated parameters of each drug and the magnitude of DDIs with probe drugs were in agreement with reference values. Darunavir/ritonavir increased erlotinib and gefitinib exposure, while efavirenz and etravirine decreased erlotinib and gefitinib concentrations. Based on our predictions, dose-adjustment strategies may consist of once-daily dosing erlotinib at 25 mg and gefitinib at 125 mg with darunavir/ritonavir; or erlotinib at 200 mg and gefitinib at 375 mg with etravirine. The interaction with efavirenz was not overcome even after doubling erlotinib or gefitinib doses. Conclusions: PBPK models predicted the in vivo pharmacokinetics of erlotinib, gefitinib and the antiretrovirals darunavir/ritonavir, efavirenz and etravirine, and the DDIs between them. The simulated dose-adjustments may represent valuable strategies to optimize antineoplastic therapy in HIV-infected patients. Introduction WidespreaduseofcombinationART(cART)hasbeenfollowedbya marked decrease in the mortality associated with HIV infection. 1 At the same time, however, the incidence of various comorbidities, including non-AIDS-defining cancers (NADCs), seems to have risen. 2–6 Several studies have shown that early and effective cART during chemotherapy increases survival rates, 7–10 andinitiationormaintenance of cART is currently recommended for HIV-infected patients with cancer. However, concomitant administration of cART and chemotherapy in clinical practice may be challenging due to the risk of overlapping toxicities and the potential for difficult-to-manage drug–drug interactions between antineoplastic and antiretroviral drugs. Many antineoplastics are substrates of metabolizing enzymes and drug transporters that can be inhibited or induced by antiretrovirals. This may lead to substantial changes in drug exposure, which, in turn, may result in increased toxic effects or in decreased efficacy of chemotherapy. Since ritonavir is a potent inhibitor of both cytochrome P450 enzymes and drug transporters involved in the disposition of V CThe Author 2016. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: [email protected]. 805 J Antimicrob Chemother 2017; 72: 805–811 doi:10.1093/jac/dkw485 Advance Access publication 13 December 2016 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023 numerous drugs, 11,12 patients taking boosted-cART regimens are those at higher risk for drug interactions with antineoplastics. 9,13,14 On the other hand, some NNRTIs may induce cytochrome P450 enzymes 15 and could potentially reduce the exposure, and consequently the efficacy, of certain chemotherapy drugs. Despite the potential relevance of this issue, clinical data on drug–drug interactions between antineoplastic and antiretroviral drugs are very scant and ethical considerations limit the possibility of running formal clinical trials specifically aimed at addressing pharmacokinetic interactions between cART and chemotherapy. Physiologically based pharmacokinetic (PBPK) modelling is a modelling approach that can be useful to predict drug–drug interactions in the absence of clinical data. This approach combines in vitro data (e.g. physicochemical characteristics, intrinsic clearance and permeability) with system data that describes the population of interest (e.g. demographic, anatomical and physiological characteristics) to simulate the pharmacokinetics of drugs. 16 This strategy is being widely used for drug development in different therapeutic areas and it has been considered by regulatory agencies as an alternative for exploring drug–drug interaction potential between a substrate drug and an interacting drug. 17 Erlotinib and gefitinib are two tyrosine kinase inhibitors (TKIs) with antiangiogenic activity which are used for the treatment of patients with several types of cancer, 18,19 including those with non-small cell lung cancer, 20,21 which is one of the most common NADCs observed in HIV-infected patients. 5,13,22–25 Although erlotinib and gefitinib may be attractive treatment options due to their oral bioavailability and relatively favourable safety profile compared with cytotoxic drugs, their use is not exempt from possible adverse effects, which may be dose limiting. 26–28 Additionally, both erlotinib and gefitinib are substrates for different isoenzymes of the cytochrome P450 system, primarily CY3A4, 18,19 which provides a rationale for significant drug interactions with antiretrovirals. Even so, clinical data on drug–drug interactions between these antineoplastics and antiretroviral agentsare lacking. The aim of this study was, therefore, to predict the magnitude of drug–drug interactions of erlotinib and gefitinib with a ritonavirboosted PI (darunavir/ritonavir) as well as with two NNRTIs (efavirenz and etravirine) simulating virtual clinical trials through PBPK modelling. The robustness of this approach was assessed by comparing the magnitude of simulated drug–drug interactions using two probe drugs for CYP3A4 (midazolam and maraviroc) with that observed in clinical trials. Finally, the validated model was used to simulate erlotinib and gefitinib dose adjustments to achieve therapeutic drug exposure when these agents are coadministered with antiretrovirals. 26–28 Methods A cohort of 50 healthy volunteers was generated. All organs and tissues were represented as individual compartments. Assumptions in the model were: (i)instantdistribution of thedrugs (well-stirred model); (ii) no drug absorption from the colon; and (iii) drug transport into tissues was blood flow limited. The PBPK model was designed using Simbiology (Matlab, version R2013b). System parameters Virtual healthy volunteers were generated using a population physiology model which provides a statistical description of the physiological and anatomical characteristics in the human population. 29 The physiology model reflects a population (0.5 proportion females) in the age range 20–50 years (mean age 36.5 613years), with average height of 1.7560.17 m and average weight of 76.3614.6 kg. Individual values of age, height and body weight were used for the calculation of organ and tissue volumes through allometric equations. 29 Additionally, systemic blood circulation was simulated considering the cardiac output and regional blood flows as previously described. 29 BasedonthestudyofYuet al., 30 oral absorption was simulated by means of a compartmental absorption and transit model. Human effective permeability was evaluated using the apparent permeability in Caco-2 cell monolayer or based on the polar surface area (PSA) and hydrogen bond donor (HBD) values. 31 Intestinal and first-pass hepatic metabolism were evaluated considering the in vitro intrinsic clearance of each enzyme metabolizing the drug and the abundance of each enzyme in both intestinal and hepatic tissue. 32,33 Hepatic clearance was simulated based on intrinsic clearance data according to Obach et al. 34 Enzyme inhibition in the gut and in the liver was described according to a mechanism-based inhibition model equation, using total concentrations of the inhibitor in the gut for gut metabolism and the free concentrations of the inhibitor in the liver for liver metabolism. Induction of enzyme expression in the gut and in the liver was expressed according to a maximum effect model equation, using total concentrations of the inducer in both cases. The volume of distribution was simulated by calculating the plasma-totissue ratio foreach organ, according to Poulin and Theil. 35 Drug parameters In vitro data describing physicochemical characteristics of erlotinib, gefitinib, ritonavir, darunavir, efavirenz and etravirine are summarized in Table S1 (available as Supplementary data at JAC Online). Similarly, data describing the metabolism of the drugs by different enzyme isoforms were available in the literature and intrinsic clearance values for these metabolic processes were included in the model as shown in Table S2. Simulation design First, steady-state drug concentrations in plasma were simulated in a virtual population of 50 individuals receiving 150 mg of erlotinib once daily, 250 mg of gefitinib once daily, 800/100 mg of darunavir/ritonavir once daily, 600 mg of efavirenz once daily or 200 mg of etravirine twice daily. Second, to facilitate validation of the model, the inhibitory and inductive potential of darunavir/ritonavir, efavirenz and etravirine was evaluated with probe drugs for CYP3A4 (midazolam and maraviroc) and simulated interactions were compared with data from drug–drug interaction studies available in the literature. Finally, the validated model was used to simulate interactions between erlotinib and gefitinib and antiretroviral drugs, as well as plausible erlotinib and gefitininb dose adjustments to establish doses attaining comparable exposure to that in the absence of antiretrovirals. Results Simulated pharmacokinetics of erlotinib, gefitinib, darunavir, ritonavir, efavirenz and etravirine are shown in Table 1.The pharmacokinetic parameters of each drug were comparable to clinical data. Interactions with reference probes Since ritonavir is a much more potent CYP3A4 inhibitor than darunavir, CYP3A4 inhibition by darunavir/ritonavir was described according to a competitive antagonistic model in which darunavir Molt oet al. 806 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023 caused minimal additional CYP3A4 inhibition when ritonavir was present. 36 As shown in Figure 1, the results of simulated trials of drug interactions between darunavir/ritonavir, efavirenz and etravirine with midazolam and maraviroc gave results comparable to those in the literature. The simulated interaction between ritonavir and midazolam (400 mg twice daily ritonavir plus a single oral dose of 2 mg midazolam) resulted in an increase in the midazolam area under the curve (AUC 0-inf ) by 10.4-fold, compared with the reported 8.14-fold in a clinical study. 37 Conversely, the simulated interaction between efavirenz (600 mg once daily) and maraviroc (100 mg twice daily) resulted in a 65% decrease in the AUC 0-12 of maraviroc, compared with 51% in the clinical study by Abel et al. 38 Finally, the interaction between etravirine and maraviroc has been evaluated by Kakuda et al. 39 who reported a 53% decrease in maraviroc AUC 0-12 when maraviroc (300 mg twice daily) was given with etravirine (200 mg twice daily), while the corresponding simulation resulted in a 49% decrease. Interactions with antineoplastics Darunavir/ritonavir caused a profound decrease in erlotinib firstpass metabolism and systemic clearance. This had a major effect on erlotinib pharmacokinetics, increasing AUC 0–24 and maximum plasma concentration (C max ) 6.8and 5.4-fold, respectively. A dose reduction of erlotinib was simulated and a daily dose of 25mg was predicted to overcome the effect of darunavir/ritonavir on erlotinib disposition (Table S3 and Figure 2a). At the standard regimen (150mg once daily), the simulated AUC 0–24 of erlotinib was equal to 32.19 (21.60–46.55) mgh/L and the interaction with darunavir/ritonavir caused an increase to 225.83 (55.66–406.84) mgh/L. The simulated dose reduction to 25mg once daily gave a median (range) erlotinib AUC 0–24 of 38.69 (17.27–55.91) mgh/L. Similarly, darunavir/ritonavir increased gefitinib AUC 0–24 and C max 5.5and 4.1-fold, respectively. It is noteworthy that gefitinib AUC 0–24 and C max were still increased by 2.79and 2.05-fold, respectively, after halving the gefitinib dose to 125 mg once daily compared with standard dosing without darunavir/ritonavir (Table S3 and Figure 2b). Even so, gefitinib individual concentrations in this scenario were considered to be within the therapeutic range and this dose adjustment was considered appropriate. 28 Conversely to darunavir/ritonavir, efavirenz and etravirine induced the metabolism of erlotinib and gefitinib, with a subsequent decrease in drug exposure. Such a decrease was particularly marked in the case of efavirenz and it was not overcome even after doubling the dose of either erlotinib or gefitinib (Table S3 and Figure 2c and d). In the case of etravirine, simulated dose increases of erlotinib and gefitinib showed that a daily dose of 200 mg erlotinib and 375 mg gefitinib was sufficient to provide pharmacokinetic parameters comparable to those obtained with standard dosing without etravirine (Table S3 and Figure 2e and f). Discussion The developed PBPK model described the pharmacokinetics of erlotinib, gefitinib, darunavir/ritonavir, efavirenz and etravirine, Table 1. Simulated drug pharmacokinetics compared with clinical data from the literature AUC 0-s (mgh/L) C max (mg/L) C trough (mg/L) Erlotinib observed 26 38.4629.5 2.1261.5 1.260.62 simulated 32.1966.88 1.7660.31 0.9560.27 Gefitinib observed 52 8.1 0.43 simulated 9.9762.11 0.6260.10 0.3360.08 Darunavir observed 53 83.99627.12 6.6361.68 1.8461.05 simulated 97.38640.36 6.3061.89 1.8661.55 Ritonavir observed 41 6.662.4 0.8460.39 0.0860.04 simulated 9.3564.66 0.7160.19 0.1360.16 Efavirenz observed 43 58.09623.05 4.0761.17 1.7761.01 simulated 72.66636.65 3.4761.56 2.6361.48 Etravirine observed 39 8.84 1.00 0.51 simulated 8.5962.25 0.7760.19 0.6660.18 AUC 0-s , area under the plasma concentration–time curve to the end of the dosing period; C max , maximum plasma concentration; C trough , trough plasma concentration. Dosage: erlotinib, 150 mg every 24 h; gefitinib, 250 mg every 24 h; darunavir/ritonavir, 800/100 mg every 24 h; efavirenz, 600 mg every 24 h; and etravirine, 200 mg every 12 h. 110 1 10 EFV-MVC 3 8 ETR-MVC 3 9 DRV/RTV-MDZ 3 7 0.5 0.5 Observed AUC ratio Simulated AUC ratio Figure 1. Scatter plot representing the AUC ratio (AUC with inhibitor or inducer /AUC without ) resulting from simulated and clinical drug interaction studies between reference CYP3A4 probe drugs and darunavir/ritonavir, efavirenz, or etravirine. The continuous line represents the line of identity and the broken lines represent a 2-fold difference from the observed values. AUC, area under the plasma concentration–time curve; MDZ, midazolam; MVC, maraviroc; DRV/RTV, darunavir/ritonavir; EFV, efavirenz; ETR, etravirine. PBPK modelling to simulate drug–drug interactions JAC 807 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023 and their interactions. The clarification of the magnitude of drug–drug interaction between antiretroviral and antineoplastic drugs is relevant since the co-administration of both treatments is becoming relatively common in HIV-infected patients with cancer 7–10 and changes in drug exposure may influence treatment outcomes. Virtual clinical trials were simulated and the extent of variation of the pharmacokinetic parameters of antineoplastics was investigated, identifying safer therapeutic options in relation to the risk of drug–drug interactions. This simulation approach can be viewed as a paradigmatic example demonstrating that PBPK modelling can be used to investigate relevant questions in difficult-to-explore clinical scenarios and to gain a better understanding of the molecular mechanisms underpinning drug disposition. 0 4 8 12 16 20 24 0 5 10 15 Time (h) Time (h) Time (h) Time (h) Erlotinib (mg/L)Erlotinib (mg/L) ERL 150 mg ERL 150 mg + DRV/RTV ERL 25 mg + DRV/RTV (a) 0 1 2 3 ERL 150 mg ERL 150 mg + EFV ERL 300 mg + EFV (c) 0 1 2 3 ERL 150 mg ERL 150 mg + ETR ERL 200 mg + ETR (e) 0 4 8 12162024 0 4 8 12 16 20 24 0 4 8 12 16 20 24 Time (h) Time (h) 0 4 8 12 16 20 24 0 4 8 12 16 20 24 0 1 2 3 Gefitinib (mg/L) Erlotinib (mg/L) Gefitinib (mg/L) Gefitinib (mg/L) GEF 250 mg GEF 250 mg + DRV/RTV GEF 125 mg + DRV/RTV (b) 0 0.2 0.4 0.6 0 0.2 0.4 0.6 GEF 250 mg GEF 250 mg + EFV GEF 500 mg + EFV (d) GEF 250 mg GEF 250 mg + ETR GEF 375 mg + ETR (f) Figure 2. Mean erlotinib and gefitinib concentration–time profiles in the absence and in the presence of antiretroviral drugs. ERL, erlotinib; GEF, gefitinib; DRV/RTV, darunavir/ritonavir; EFV, efavirenz; ETR, etravirine. Molt oet al. 808 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023 Proper use of PBPK models to simulate drug–drug interactions requires deep knowledge of the physicochemical and metabolic properties of the drugs, including CYP inhibition and induction. Data describing the metabolism of drugs by different isoforms of the cytochrome P450 system were available from the literature. Erlotinib and gefitinib are metabolized primarily by CYP3A4 and to a lesser extent by other isoenzymes. 40 Ritonavir and darunavir are also primarily metabolized by the CYP3A isozyme family. 41,42 Similarly, CYP2B6 and CYP3A4 are the major isozymes responsible for efavirenz metabolism, while etravirine is metabolized by CYP3A4 and CYP2C19. 43–45 According to FDA classification (moderate and strong inhibition: >2-fold but <5-fold increase in AUC,>5-fold increase in AUC, respectively; moderate and strong induction: 50%–80% decrease in AUC,>80% decrease in AUC, respectively), 17 the magnitude of the simulated drug–drug interactions with erlotinib or gefitinib was moderate to strong. These results are in agreement with previous preclinical studies, 46,47 and can be explained because both erlotinib and gefitinib are mainly metabolized by CYP3A4, 40 which is strongly impacted by ritonavir and efavirenz as well as, to a lesser extent, by etravirine. 27–39,41,43–45 Ritonavir is a strong inhibitor and a weak inducer of CYP3A4, while darunavir is a weak inhibitor and inducer of CYP3A4. 41,42,48 When considering drug interactions with darunavir/ritonavir, ritonavir was the main perpetrator and darunavir was considered to cause minimal additional CYP3A4 inhibition when ritonavir was present. This assumption was based on data by German et al. 36 who showed no additional CYP3A4 inhibition by ketoconazole to that caused by ritonavir. According to prescribing information, caution is advised when erlotinib needs to be co-administered with strong CYP3A4 inhibitors. 18 Such a recommendation is based on data showing an increase in erlotinib AUC of 80% when it was co-administered with ketoconazole. However, the increase in erlotinib concentrations caused by ritonavir in our study was much more pronounced than initially expected. This finding, together with the fact that erlotinib is dosed at its maximum tolerated dose due to limiting toxicities, 26,27 makes it essential to reduce erlotinib dose in this scenario to as low as 25 mg once daily or to look for alternative treatment options. Gefitinib has a wider therapeutic range than erlotinib and the approved dose of gefitinib is only one-third of its maximum tolerated dose. 19,28 Thus, although gefitinib concentrations in the presence of ritonavir were still high after reducing the standard dose by 50%, individual concentrations were considered to be within the therapeutic range, 28 and such a dose adjustment was considered appropriate. At drug concentrations observed in clinical practice, CYP3A4 induction is more pronounced with efavirenz than with etravirine, 43–45 as reflected by the magnitude of their interactions. Our data showed a particularly marked decrease in erlotinib and gefitinib exposure when they were administered with efavirenz. Moreover, such interactions were not overcome even after doubling the erlotinib or gefitinib doses, making these combinations not recommended in clinical practice, especially if an alternative is available. Etravirine is a weaker inducer of CYP3A4 and its impact on erlotinib and gefitinib pharmacokinetics was low. In this case, we were able to identify potential dose adjustments to provide therapeutic concentrations of erlotinib and gefitinib in the presence of etravirine. Gefitinib and erlotinib are substrates of drug transporters including ABCB1, BCRP and MRPs, 47 whose activity may be influenced by ritonavir, efavirenz or etravirine. 49,50 However, both erlotinib and gefitinib are very potent inhibitors of these transporters and no additional inhibition was observed after co-incubation with other paradigm inhibitors such as cyclosporine A (an ABCG2 inhibitor), KO-143 (an ABCG2 inhibitor) and MK-571 (an inhibitor of ABCCs). 51 Despite the potential relevance of transporter activity in erlotinib and gefitinib disposition, we considered that there are still too many gaps in the knowledge of these processes to incorporate them into the current model. Although this might be considered as a limitation of the present study, our modelling approach proved to be robust. The simulated pharmacokinetics of erlotinib, gefitinib, darunavir/ritonavir, efavirenz and etravirine were in good accordance with previously described clinical data. Additionally, the inhibitory and inductive effects of the perpetrator drugs were validated by comparing the magnitude of simulated drug–drug interactions using probe drugs with that observedin clinical studies. The absence of comparisons using other boosted PIs (e.g. atazanavir/ritonavir) may also be considered as a potential limitation of this study. However, as mentioned above, at clinical concentrations ritonavir causes almost complete inhibition of CYP3A4, with no additional inhibition by the addition of a second drug, 36 making our results potentially applicable to the scenario of atazanavir/ritonavir-containing cART. Finally, we used a healthy patient population profile instead of HIV-infected or cancer patient profiles when performing the simulations. Pharmacokinetic differences between healthy volunteers and patients are well known due to different tissue composition, altered protein binding, etc. Consequently, discrepancies in drug exposure and also in the magnitude of drug–drug interactions between studies performed in healthy volunteers or in patients have been reported in some cases. This consideration makes a clinical validation of our results highly desirable. In summary, we developed a PBPK model that predicted the pharmacokinetics of erlotinib, gefitinib, darunavir/ritonavir, efavirenz and etravirine, and their interactions. The clarification of the magnitude of drug–drug interactions between antiretrovirals and chemotherapeutic drugs is relevant since it may influence treatment outcome. The model presented here provides a rational platform both to optimize antineoplastic therapy in HIV patients receiving these key antiretrovirals as well as to inform the design for a clinical drug interaction study that may save time and resources, while the optimal dose is determined empirically. Funding This work was supported by Lluita contra la Sida Foundation. M. V. is supported by FIS through grant CP04/00121 from the Spanish Health Department in collaboration with Institut de Recerca de l’Hospital de la Santa Creu i Sant Pau, Barcelona, and is a member of CIBERSAM (a research network funded by the Spanish Health Ministry, Instituto de Salud Carlos III). Transparency declarations J. M., D. B. and B. C. have received research funding, consultancy fees, and lecture sponsorships from and have served on advisory boards for various laboratories (MSD, AbbVie, Boehringer Ingelheim, Gilead Sciences, ViiV PBPK modelling to simulate drug–drug interactions JAC 809 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023 Healthcare, Janssen-Cilag and Bristol-Myers Squibb). A. O. has received research funding from MSD, ViiV Healthcare, Pfizer and AstraZeneca. M. S. has received research funding from Janssen and ViiV Healthcare. All other authors: none to declare. Supplementary data Tables S1–S3 are available as Supplementary data at JAC Online (http://jac. oxfordjournals.org/). References 1Clifford GM, Polesel J, Rickenbach M et al. Cancer risk in the Swiss HIV Cohort Study: associations with immunodeficiency, smoking, and highly active antiretroviral therapy. J Natl Cancer Inst 2005; 97: 425–32. 2Deeken JF, Tjen-A-Looi A, Rudeck MA et al. The rising challenge of nonAIDS-defining cancers in HIV-infected patients. Clin Infect Dis 2012; 55: 1228–35. 3Shiels MS, Pfeiffer RM, Gail MH et al. Cancer burden in the HIV-infected population in the United States. JNatlCancerInst2011; 103: 753–62. 4Frisch M, Biggar RJ, Engels EA et al. Association of cancer with AIDS-related immunosuppression in adults. JAMA 2001; 285: 1736–45. 5Hessol N, Mart ınez-Maza O, Levine AM et al. Lung cancer incidence and survival among HIV-infected and uninfected women and men. AIDS 2015; 29: 1183–93. 6Engels EA, Pfeiffer RM, Goedert JJ et al. Trend in cancer risk among people with AIDS in the United States 1980-2002. AIDS 2006; 20: 1645–54. 7Lavole A, Chouaid C, Baudrin L et al. Effect of highly active antiretroviral therapy on survival of HIV-infected patients with non-small-cell lung cancer. Lung Cancer 2009; 65: 345–50. 8Navarro JT, Ribera JM, Oriol A et al. Favourable impact of virological response to highly active antiretroviral therapy on survival in patients with AIDS-related lymphoma. Leuk Lymphoma 2002; 43: 1837–42. 9Torres HA, Rallapalli V, Saxena A et al. Efficacy and safety of antiretrovirals in HIV-infected patients with cancer. Clin Microbiol Infect 2014; 20: 672–9. 10 Torres HA, Mulanovich V. Management of HIV infection in patients with cancer receiving chemotherapy. Clin Infect Dis 2014; 59: 106–14. 11 King JR, Wynn H, Brundage R et al. Pharmacokinetic enhancement of protease inhibitor therapy. Clin Pharmacokinet 2004; 43: 291–310. 12 Renjifo B, van Wyk J, Salem AH et al. Pharmacokinetic enhancement in HIV antiretroviral therapy: a comparison of ritonavir and cobicistat. AIDS Rev 2015; 17: 37–46. 13 Bower M, Powels T, Stebbing J et al. Potential antiretroviral drug interactions with cyclophosphamide, doxirrubicin, and etoposide. JClinOncol 2005; 23: 1328–9. 14 Leveque D, Santucci R, Pavillet J et al. Paralytic ileus possibly associated with interaction between ritonavir/lopinavir and vincristine. Pharm World Sci 2009; 31: 619–21. 15 Hariparsad N, Nallani SC, Sane RS et al. Induction of CYP3A4 by efavirenz in primary human hepatocytes: comparison with rifampin and phenobarbital. J Clin Pharmacol 2004; 44: 1273–81. 16 Siccardi M, Rajoli RKR, Curley P et al. Physiologically based pharmacokinetic models for the optimization of antiretroviral therapy: recent progress and future perspective. Future Virol 2013; 8: 871–90. 17 US FDA. Drug Development and Drug Interactions. http://www.fda.gov/ Drugs/DevelopmentApprovalProcess/DevelopmentResources/ DrugInteractionsLabeling/ucm080499.htm. 18 Erlotinib (Tarceva). Product Information. http://www.ema.europa.eu/ docs/en_GB/document_library/EPAR_-_Product_Information/human/ 000618/WC500033994.pdf. 19 Gefitinib (Iressa). Product Information. http://www.ema.europa.eu/docs/ en_GB/document_library/EPAR_-_Product_Information/human/001016/ WC500036358.pdf. 20 Okuma Y, Hosomi Y, Imamura A. Lung cancer patients harboring epidermal growth factor receptor mutation among those infected by human immunodeficiency virus. Onco Targets Ther 2015; 8: 111–5. 21 Erikson TM, Koeppe JR, Miller YE et al. Bronchioloalveolar carcinoma presenting as chronic progressive pulmonary infiltrates in a woman with HIV: a diagnosis worth making. JThoracOncol2008; 3: 1353–5. 22 Chaturvedi AK, Pfeiffer RM, Chang L et al.Elevatedriskoflungcancer among people with AIDS. AIDS 2007; 21: 207–13. 23 Sigel K, Wisnivesky J, Gordon K et al. HIV as an independent risk factor for incident lung cancer. AIDS 2012; 26: 1017–25. 24 Bearz A, Vaccher E, Martellotta F et al. Lung cancer in HIV patients: the GICAT experience. Eur Rev Med Pharmacol Sci 2014; 18: 500–8. 25 Bower M, Powles T, Nelson M et al. HIV-related lung cancer in the era of highly active antiretroviral therapy. AIDS 2003; 17: 371–5. 26 Hidalgo M, Siu LL, Nemunaitis J et al. Phase I and pharmacologic study of OSI-774, an epidermal grow factor receptor tyrosine kinase inhibitor, in patients with advanced solid malignancies. JClinOncol2001; 19: 3267–79. 27 Rudin CM, Liu W, Desai A et al. Pharmacogenomic and pharmacokinetic determinants of erlotinib toxicity. JClinOncol2008; 26: 1119–27. 28 Nakagawa K, Tamura T, Negoro S et al. Phase I pharmacokinetic trial of the selective oral epidermal growth factor receptor tyrosine kinase inhibitor gefitinib (‘Iressa’, ZD1839) in Japanese patients with solid malignant tumors. Ann Oncol 2003; 14: 922–30. 29 Bosgra S, van Eijkeren J, Bos P et al. An improved model to predict physiologically based model parameters and their inter-individual variability from antropometry. Crit Rev Toxicol 2012; 42: 751–67. 30 Yu LX, Amidon GL. A compartmental absorption and transit model for estimating oral drug absorption. Int J Pharm 1999; 186: 119–25. 31 Gertz M, Harrison A, Houston JB et al. Prediction of human intestinal firstpass metabolism of 25 CYP3A substrates from in vitro clearance and permeability data. Drug Metab Dispos 2010; 38: 1147–58. 32 Houston JB. Utility of in vitro drug metabolism data in predicting in vivo metabolic clearance. Biochem Pharmacol 1994; 47: 1469–79. 33 Ohtsuki S, Schaefer O, Kawakami H et al. Simultaneous absolute protein quantification of transporters, cytochromes P450, and UDPglucuronosyltransferases as a novel approach for the characterization of individual human liver: comparison with mRNA levels and activities. Drug Metab Dispos 2012; 40: 83–92. 34 Obach RS. Prediction of human clearance of twenty-nine drugs from hepatic microsomal intrinsic clearance data: An examination of in vitro half-life approach and nonspecific binding to microsomes. Drug Metab Dispos 1999; 27: 1350–9. 35 Poulin P, Theil FP. Prediction of pharmacokinetics prior to in vivo studies. 1. Mechanism-based prediction of volume of distribution. JPharmSci2002; 91: 129–56. 36 German P, Mathias A, West S et al. Evaluation of ritonavir-boosted elvitegravir PK upon co-administration with a second potent CYP3A inhibitor, ketoconazole. In: Abstracts of the Eleventh International Workshop on Clinical Pharmacology of HIV Therapy, Sorrento, Italy, 2010. Abstract 48. 37 Kirby BJ, Collier AC, Kharasch ED et al. Complex drug interactions of HIV protease inhibitors 1: inactivation, induction, and inhibition of cytochrome P450 3A by ritonavir or nelfinavir. Drug Metab Dispos 2011; 39: 1070–8. Molt oet al. 810 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023 38 Abel S, Jenkins TM, Whitlock LA et al. Effects of CYP3A4 inducers with and without CYP3A4 inhibitors on the pharmacokinetics of maraviroc in healthy volunteers. Br J Clin Pharmacol 2008; 65 Suppl 1: 38–46. 39 Kakuda TN, Abel S, Davis J et al. Pharmacokinetic interactions of maraviroc with darunavir-ritonavir, etravirine, and etravirine-darunavir-ritonavir in healthy volunteers: results two drug interaction trials. Antimicrob Agents Chemother 2011; 55: 2290–6. 40 Li J, Zhao M, He P et al. Differential metabolism of gefitinib and erlotinib by human cytochrome P450 enzymes. Clin Cancer Res 2007; 13: 3731–7. 41 Ritonavir (Norvir). Product Information. http://www.ema.europa.eu/docs/ en_GB/document_library/EPAR_-_Product_Information/human/000127/ WC500028728.pdf. 42 Darunavir (Prezista). Product Information. http://www.ema.europa.eu/ docs/en_GB/document_library/EPAR_-_Product_Information/human/ 000707/WC500041756.pdf. 43 Efavirenz (Sustiva). Product Information. http://www.ema.europa.eu/ docs/en_GB/document_library/EPAR_-_Product_Information/human/ 000249/WC500058311.pdf. 44 Etravirine (Intelence). Product Information. http://www.ema.europa.eu/ docs/en_GB/document_library/EPAR_-_Product_Information/human/ 000900/WC500034180.pdf. 45 Yanakis LJ, Bumpus NN. Biotransformation of the antiretroviral drug etravirine: metabolite identification, reaction phenotyping, and characterization of autoinduction of cytochrome P450-dependent metabolism. Drug Metab Dispos 2012; 40: 803–14. 46 Pillai VC, Venkataramanan R, Parise RA et al. Ritonavir and efavirenz significantly alter their metabolism of erlotinib—an observation in primary cultures of human hepatocytes that is relevant to HIV patients with cancer. Drug Metab Dispos 2013; 41: 1843–51. 47 Deeken JF, Beumer JH, Anders NM et al. Preclinical assessment of the interactions between the antiretroviral drugs ritonavir and efavirenz, and the tyrosine kinase inhibitor erlotinib. Cancer Chemother Pharmacol 2015; 76: 813–9. 48 Rittweger M, Arasteh K. Clinical pharmacokinetics of darunavir. Clin Pharmacokinet 2007; 46: 739–56. 49 Gupta A, Zhang Y, Unadkat JD et al. HIV protease inhibitors are inhibitors but not substrates of the human breast cancer resistance protein (BCRP/ ABCG2). J Pharmacol Exp Ther 2004; 334–41. 50 Profit L, Eagling VA, Back JD. Modulation of P-glycoprotein function in human lymphocytes and Caco-2 cell monolayers by HIV-1 protease inhibitors. AIDS 1999; 13: 1623–7. 51 Lainey E, Se´bertM,The´pot Set al. Erlotinib antagonizes ABC transporters in acute myeloid leukemia. Cell Cycle 2012; 11: 4079–92. 52 Adjei AA, Molina JR, Mandrekar SJ et al. Phase I trial of sorafenib in combination with gefitinib in patients with refractory or recurrent non-small cell lung cancer. Clin Cancer Res 2007; 13: 2684–91. 53 Molt o J, Valle M, Ferrer E et al. Reduced darunavir dose is as effective in maintaining HIV suppression as the standard dose in virologically suppressed HIV-infected patients: a randomized clinical trial. JAntimicrobChemother 2015; 70: 1139–45. PBPK modelling to simulate drug–drug interactions JAC 811 Downloaded from https://academic.oup.com/jac/article/72/3/805/2633725 by Universite de Geneve user on 06 December 2023