In Vitro In Vivo Extrapolation (IVIVE): Why It Is Not As Easy As - PowerPoint PPT Presentation

EMA Workshop on MPS - 2017 In Vitro In Vivo Extrapolation (IVIVE): Why It Is Not As Easy As You May Think Amin Rostami Professor of Systems Pharmacology University of Manchester, UK & Chief Scientific Officer & Senior Vice President of R&D Certara , Princeton, USA

Our Experience w ith MPS: LiverChip TM ( Ayşe Ufuk, Tom De Bruyn )  Perfusion culture plate incorporates integrated scaffold permitting formation of 3D liver microtissues that resembles the architecture of a liver sinusoid

Initial LiverChip Optimisation Studies  Albumin secretion in hepatocytes cultured in LiverChip TM is stable for up to 9 days of culture time  Urea medium concentrations are higher compared to static 2D cultures  Urea synthesis decreases with time  Effect of culture time on enzyme activity was evaluated  CYP2C9 activity stable – no significant difference in tolbutamide depletion and 4OH-tolbutamide formation on day 3-4 and days 6-7  Inter-day variability based on tolbutamide depletion as a marker was evaluated  Approximately 40% variation in tolbutamide clearance was observed Ufuk et al, manuscript in preparation Other Team Members: Tom De Bruyn, Michiharu Kageyama, Alex Galetin, Brian Houston, David Hallifax 3 3

IVIVE: Major Input for PBPK (and any other QSP) Models In Vitro to In Vivo Extrapolation

PBPK: Typical View = Nothing New ! Lung Typical View : Generic Adipose Venous Blood Describing the C-T profiles Arterial Blood Bone based on physiological Brain know ledge of the flow s Heart and partitions Kidney Muscle Skin Liver Spleen Portal Vein Gut Nothing New : IV PO Teorell, T. Studies on the diffusion effect upon ionic distribution: II. experiments on ionic accumulation. J. Gen. Physiol. 21, 107–122 (1937)

How it is done? Integrating system information • Replacement and additional organ Permeability-limited models are available for the intestine, liver, kidney, brain and lung. • Transport across a membrane is often defined as Perfusion Limited • But we now define uptake/efflux into/out of selected organs as Permeability Limited

Adoption in Industrial Scale From Academic Nicety to Industrial Necessity MHRA (PSP 2015) FDA (PSP 2015)

PBPK under the Umbrella of Systems Pharmacology Multi-Level Hybrid Models: The Framew ork Direct Dose Response for Capturing, Retaining PD-DDI, Finney, Loewe, Bliss, Hewlett&Plackett., Greco, Vølund...... & Re-using the Available Response surfaces Generalized Linear models (GLM) Systems Know ledge at a PKPD... Hill, Indirect Physiological Response, Given Time Disease Progression, Cell Growth/Death GLM, Survival Analysis, PKPD-DDI Receptor binding...in vitro...system response Equilibrium binding, specific receptors, Operational Agonism T DT Receptor states, G-Proteins & ternary complexes ... signalling T1G DT1G Ion channel kinetic models ... T2G DT2G T1 DT1 T3G DT3G T2 DT2 T3 DT3 Specific Diseases, .........safety ....QTc ...or Systems Biology X-fertilization ... Network response motifs, Combinatorial targets Hill in genetic regulatory networks

Reduction in Traditional Use of Animal One for Man, Two for Horse, G. Carson, Bramhall House, New York, 1961

Interspecies Differences in Metabolising Enzymes A major component of PBPK is information on metabolism. Different P450 Mediated Activities in 4 Species 3 human nmol/(mg/min) 2.5 Metabolite horse 2 dog 1.5 cat 1 0.5 0 1A1/2 1A1/2 2A6 2A6 2C8/9 2C8/9 2C19 2C19 2D6 2D6 2 E 1 2 E 1 3A4 3A4 Specific Probe Compound for Human P450 Enzyme Chauret et al ., 1997

Modelling and Simulation A related area in modernizing clinical trials has been the development and application of quantitative pharmacometric predictive models to support regulatory decision making. Modeling and simulation (M/S) tools for drug exposure and its response have been useful in both pre- and postmarket settings when questions related to safety and efficacy of therapeutic products arise. Some recent examples where M/S has served as a useful predictive tool include dose selection for pivotal trials, dosing in select populations such as pediatrics, optimization of dose and dosing regimen in a subset patient population, prediction of efficacy and dosing in an unstudied patient population in clinical trials, characterizing exposure and dose-related QT interval prolongation, and using physiologically based pharmacokinetic

Path to Success in Using PBPK-IVIVE and Virtual Humans Path (I)  Refining In Vitro Tests for Quantitative IVIVE Path (II)  Providing & Integrating System Information Path (III)  Transparent Methods and Case Examples Path (IV)  Showing Value & Re-Engineering Practices

The Debate at the Time: Which In Vitro System to Use? Human Liver Microsomes (HLM) Recombinantly expressed system (rhCYP) Hepatocytes Fractionation & Isolation Intact cells containing full of Enriched Organelles complement of drug metabolising enzymes Differences in intrinsic activity between rhCYP & HLM Infrequent supply / cost Donor variability

How Representative Is My HLM/Hepatocyte? - High degree of lot-to- lot consistency for CYP and UGT activity - Representation of the “average patient” and known CYP polymorphisms

Determination of Intrinsic Clearance: Right Units CL In Vitro CLu int CLu = int int fu inc μ l / min / per functional unit of system In vitro system [dA/dt] (pmol/min/mg microsomal protein) HLM [S] (nmol/ml) [S] = Free Substrate Concentration [dA/dt] (pmol/min/10 6 cells) HHEP [S] << Km (hopefully!) [S] (nmol/ml) [dA/dt] (pmol/min/pmol CYP isoform) rhCYP [S] (nmol/ml)

Applying Appropriate Scaling Factors in Human IVIVE In vitro CLu int per CLu int g Liver Scaling Scaling In vitro CLu int per Factor 1 Factor 2 system Liver µL.min -1 HLM MPPGL X mg mic protein µL.min -1 Liver HHEP X X HPGL Weight 10 6 cells µL.min -1 pmol P450 isoform rhCYP X MPPGL X pmol P450 isoform mg mic protein

Literature Values: Human Microsomal Protein per Gram of Liver Reports on Assessing Human MPPGL 1970 1980 1990 2000 2010 Schoene et al. [36] Beaune [53] Baarnhielm et al., [14] Knaak et al., [25] Iwatsubo et al., [5] Lipscomb et al., [30] 1972 ; 35 1982 (19) 1986 ; 77 1993 ; 7 1997a ; 52.5 2003 ; 56 Pelkonen et al. [34] Lipscomb et al., [29] Wilson et al., [42] 1973 ; 35 1998 ; 21 2003 ; 33 Hakooz et al., [20] Pelkonen et al. [35] 1974; 36 2006 ; 40 Barter et al. , In preparation ; 29 Pelkonen, [78] Galetin et al., [19] Obach et al., [32] Carlile et al., [18] Some Reports 2004 ; 40 1999 ; 77 1997 ; 45 1999 ; 50 Predicting Human Lu et al., [71] Hepatic Clearance 2006 ; 45 Howgate et al., [9] Kuperman et al., [26] 2006 ; 33 1994 ; 45 Boase & Miners [79] Houston, [11] 2002 ; 45 Bayliss et al., [16] 1994 ; 45 Mohutsky et al., [72] 1999 ; No Values 2006 ; 45 Rat MPPGL Uchaipichat et al., [70] Walker et al., [41] 2006 ; 45 Le Goff et al., [27] Li et al., [28] 1996 ; 45 2002 ; 45 2003 ; 45 Soars et al., [39] Anderson et al., [4] 2002 ; 45 2001 ; 45 No Reference? Barter et al. (2007) Current Drug Metabolism Scaling factors for the extrapolation of in vivo metabolic drug clearance from in vitro data: Reaching a consensus on values of human microsomal protein and hepatocellularity per gram of liver

Literature Values: Number of Human Hepatocytes per Gram of Liver Reports on Assessing Human HPGL (10 6 cells/g) 1970 1980 1990 2000 2010 Lipscomb et al., [29] Wilson et al., [42] Barter et al., 1998; 116 2003; 107 In preparation ; 86 Arias , [13] Some Reports Szakacs et al., [40] McGinnity et al., [31] No Reference? 1988; 120 2001; 135 2004; 120 Predicting Human Hepatic Clearance Bayliss et al., [17] Soars et al., [39] Bayliss et al., [16] Kuperman et al. [26] 1990; 120 2002 ; 120 1999 ; 120 1994 ; 120 Zuegge et al., [7] Ekins & Obach [80] Naritomi et al., [77] Bachman et al., [15] Iwatsubo et al., [5] 2001; 120 2000; 120 2003; 120 2003; 120 1997a; 120 Barter et al. (2007) Current Drug Metabolism Scaling factors for the extrapolation of in vivo metabolic drug clearance from in vitro data: Reaching a consensus on values of human microsomal protein and hepatocellularity per gram of liver

Scaling of rhCYP Data 4.3% CYP1A2 12.2% CYP isoform CYP2A6 5.0% 33.0% CYP2B6 3.8% abundance: CYP2C8 5.4% CYP2C9 CYP2C18 pmols CYP isoform CYP2C19 CYP2D6 per CYP2E1 16.5% 0.3% CYP2J2 mg of microsomal protein 0.2% CYP3A4 14.0% 3.4% 1.9% CYP3A5  Many groups use Shimada et al. (1994) values Don’t differentiate between Japanese and Caucasian  Literature review for papers reporting enzyme abundance values – Caucasian population 30-40 papers reviewed; 19-27 used for meta-analysis  Calculated weighted means, CVs and tested for homogeneity

HPGL Determination: Study by Simcyp Group (Sheffield) Number of cells 10g Liver weight (g) Centrifugation Perfusion of liver sample Human Liver Weight Counting of Calculation @ 50g to isolate with digestion media to Sample recorded cells of HPGL hepatocytes produce a homogenous suspension of hepatocytes Problem 1 Problem 2 Incomplete digestion leading to Incomplete recovery of cells incomplete release of hepatocytes following centrifugation into suspension Loss of Hepatocytes UNDERESTIMATION IN HPGL