Microdosing: Past and Future (an introduction to microdosing and where it’s heading) Dr Graham Lappin Chief Scientific Officer Xceleron Inc Insert your logo in this area then delete this text box.

Disclaimer The views and opinions expressed in the following PowerPoint slides are those of the individual presenter and should not be attributed to Drug Information Association, Inc. (“DIA”), its directors, officers, employees, volunteers, members, chapters, councils, Special Interest Area Communities or affiliates, or any organization with which the presenter is employed or affiliated. These PowerPoint slides are the intellectual property of the individual presenter and are protected under the copyright laws of the United States of America and other countries. Used by permission. All rights reserved. Drug Information Association, Drug Information Association Inc., DIA and DIA logo are registered trademarks. All other trademarks are the property of their respective owners. 2

Summary  Microdosing with PET is covered by Martin Bauer – the current presentation focuses on ‘classical pharmacokinetic’ outcomes  The definition of microdosing and its concept  How microdosing is applied  Pharmacokinetic scalability in context of drug selection  The latest applications of microdosing  Microdosing is an evolving technology 3

What is microdosing? g g A microdose “is not more than a total dose of 100 μg (or 100 th the predicted pharmacologic dose) that can be administered as a single dose or divided doses in any subject” A microdose study is one of a suit of exploratory studies “intended to be conducted in early phase-I that involve limited human exposure, have no therapeutic intent and are not intended to examine clinical tolerability” ICH M3 Regulatory Guideline 4 Microdose of investigational drug Early pharmacokinetics Microdosing has evolved since this initial definition as we will see

PK PK scalability 5 Extrapolation from animal or in vitro to human via often complex mathematical models Traditional models human to human directly reflecting PK Microdose

Microdose and Tracer Studies  Microdose and tracer studies can be confused with each other  The drug in a microdose study may be 14 C-labelled but it does not have to be  In a microdose study, the total mass of drug administered is limited to a maximum of 100 μg, irrespective of number of administrations or number of routes  In a tracer study the drug has to be isotopically labelled (usually 14 C).  The amount of drug administered can exceed 100 μg  In some cases a tracer of 14 C-drug may be administered by one route (eg intravenous) and a therapeutic dose administered by a different route (eg oral) There are some notable confusions in the literature

The origins of microdosing The technique was first proposed as a method of obtaining early PK in humans nearly 10 years ago Lappin, G., and Garner, R. C. (2003) Nat Rev Drug Discov 2,

Regulatory Guidelines  EMEA, Position Paper on Non-clinical Safety Studies to Support Clinical Trials with a Single Microdose. Position paper CPMP/SWP/2599, 23 June  Food and Drug Administration US Department of Health and Human Services Guidance for Industry Investigators and Reviewers. Exploratory IND Studies (2006)  ICH Topic M3 Note for Guidance on non-clinical safety pharmacology studies for human pharmaceuticals CPMP/ICH/286/95 (December 2009), CPMP/ICH/286/95. A single dose toxicology study, with 1,000-fold safety factor, in the rat with a 14 day follow-up with histopathology and biochemistry. (IV dose route preferred) Only requires gram amounts of non-GMP drug 8

The utility of microdosing  Microdosing is useful where a pharmacokinetic question is key to the drug’s future development  Microdosing can de-risk the selection of drug candidate in cases where traditional methods (eg allometry) do not provide reliable information  Where a choice needs to be made on the selection of a lead compound from several candidates  In risk mitigation on the choice of the most appropriate back- up compound  More recent applications such as site of action PK, drug-drug interactions and polymorphism 9

Attempted prediction of human PK by mathematical models Traditionally there are 3 main techniques of predicting human PK: Allometry In vitro-in vivo extrapolation (IVIVE) PB-PK models *Rowland & Benet (2012) J Pharm Sci 2012 Pharmaceutical Research Manufacturers Association (PhRMA) study* Blinded data from 108 drugs (only 19 given to humans IV) Outcome of human PK unknown and so bias removed Human PK predictions made using the above 3 techniques Overall conclusion was that for oral administration, the best methods predicted oral bioavailability around 45% of the time. 10

Microdose vs mathematically-based models  Direct comparisons between mathematically-based models and microdosing are difficult to make  Microdose database is small  Microdose data in public domain mainly on “old” drugs  Until the PhARMA study, modelling data were often interpreted in light of known human PK at therapeutic doses and the data fitted  Similar bias could be present in microdose data where the expected outcome was known in advance.  More microdose data needs to be published consisting of modern drugs in development 11

Summary of microdose data * being within a factor of 2 There are 44 drugs where the data are in the peered reviewed literature comparing microdose PK to therapeutic dose PK:  Oral (34) and IV (11) routes of administration  Of the oral drugs, 82% were scalable  Of the IV drugs, 100% were scalable  These are a mixture of human and animal data Although care has to be taken making the comparisons, microdosing does look promising 12 Lappin and Burt, A decade of microdosing. Where are we now and what is the future? Expert opinion in Drug Metabolism & Toxicology Due later this year

Principal published human microdose trials  CREAM Trial  Lappin et al (2006) Clin Pharmacol Ther 80,  EUMAPP  Lappin, et al (2010) Eur J Pharm Sci 40, 125–131  Lappin et al (2011) Eur J Pharm Sci 43,  NEDO  Yamane et al 2009 Drug Metab Pharmacokinet 2009; 24(4):  Tozuka et al 2010, Clin Pharmacol Ther 2010; 88(6):  Yamazaki et al (2010), J Clin Pharm Ther 2010; 35(2):

Dose-dependent PK Possible saturable kinetics Potentially any drug, if the dose can be taken up high enough, will exhibit dose-dependent kinetics The question is more whether non-linearity will be within a clinically-relevant dose range 14

Midazolam Oral microdose (100 ug) midazolam Oral 7.5 mg dose midazolam Parameter Oral dose (mg) 0.1 mg 7.5 mg t 1/ %F2223 Midazolam undergoes high first pass metabolism via CYP3A4 CYP3A4 is unlikely to be saturable at a 7.5 mg dose Lappin et al Clin Pharmacol Ther, (3): p

Propafenone Oral microdose (100 ug) propafenone Oral 150 mg dose propafenone Propafenone undergoes high first pass metabolism via CYP2D6 CYP2D6 is likely saturable at 150 mg dose Lappin, G., et al (2011) Eur J Pharm Sci 43: Parameter Oral dose 0.1 mg 150 mg t 1/ %F Drug has low bioavailability!

Bioavailability of ZK253  ZK-253 was a development drug from Schering  Allometry was highly variable (bioavailability was 27% in rat, 73% in mice, 5% in dog and 0.7% in monkey)  In Phase-I bioavailability was approximately 1%  The microdose study predicted a bioavailability of 0.16% (AMS LOQ of 300 fg/mL required) Did the microdose study under-predict the bioavailability by 6.25 fold or did it predict a fatally low bioavailability? (Lappin et al Clin Pharmacol Ther, (3): p ) 17

Accounting for saturable PK  In most cases non-linearity is due to saturation of gut wall rather than hepatic metabolism or transport  Linearity Index (LIN) calculated from dose-normalised AUC and dose/Km for 54 CYP3A4 and P-gP substrates  Combination of in vivo and in vitro data used in calculations  Predicting saturation of CYP3A4 or P-gP in gut wall when a given dose is reached  Useful approach and an interesting start but complexities can exist with absorption kinetics  Solubility and dissolution could be problematic and few BCS class II drugs have been reported for microdosing Tachibana et al (2012) Pharm Res 29:

Low dose non-linearity Possible lack of steady state 19

Warfarin Low capacity High affinity binding site and low Vol Dist Levy et al J Pharm Sci, (5): p Lack of steady state due to target mediated disposition Lappin et al Clin Pharmacol Ther, (3): p

Receptors rapidly saturated at therapeutic dose (5mg) Plasma levels fall as warfarin attaches to receptor at microdose (100 μg) Equilibrium reached warfarin cleared as it disengages from receptor over time Warfarin result explained 21

Where is microdosing heading?  Site of action PK in samples other than plasma  Eg, biopsy, CSF, specific cell types  Early read-out on metabolism, rates and routes of excretion  Gaining early PK from routes of administration other than oral eg dermal and ocular  Gaining early PK with AMS and early PD with PET  Early investigation of drug-drug interactions  Study of genetic polymorphism  Safe option to study metabolism and PK in paediatrics  Safe option to obtain metabolism and PK in the disease state (ie patient populations) 22

Early assessment as to whether the drug reaches its intended site of action Blood (and cells) Lymph Cerebrospinal fluid (CSF) Synovial fluid Biopsy Bronchoalveolar lavage (BAL) Bile (Entero-Test*) * - Guiney et al (2012) Br J Clin Pharmacol, in press Saliva, tears and nasal tissue Hair folicles Tumor (during surgery) 23

Intracellular activation of ZDV ZDV Triphosphorylation in the peripheral blood mononuclear cells (PBMCs) Plasma PK does not reflect the activity of ZDV Measurement of ZDV-TP in the PBMCs is necessary Vuong et al (2007) J Pharm Sci 97(7) Plasma 24

Study design Microdose (100 μg) 14 C-ZDV or 300 mg non-labeled ZDV per subject Blood samples taken for 2 time-points Preparation of PBMCs Separation of ZDV-TP Measurement of 300 mg dose samples with LC-MS/MS Measurement of 100 μg samples with AMS 25 Study performed in collaboration with John Hopkins/GSK

Intracellular concentrations of ZDV-TP Chen et al (2010) Clin Pharmacol Ther 88(6) Intracellular concentration of ZDV-TP (all data) LOQ: AMS 1.67 amole/sample LC-MS/MS 50 fmol/sample Dose-normalised data between 300 mg and 100 μg doses scaled well 26 AMS LC-MS/MS

Site of Action PK  Plasma PK poorly reflects the pharmacodynamics  Selection of drug candidates on plasma PK could be misleading  The ability to measure the active drug at the site of action gives a much improved correlation to efficacy  Use of 14 C enables metabolite profile as well as measurement of parent drug 27

Early read-out on metabolism Zhou et al (2009), J Clin Pharmacol. 28

Other applications of microdosing  Combination of AMS and PET in microdosing  Wagner et al (2010) Clin Pharmacokinet 50,  Early investigation of drug-drug interactions  Croft et al (2012) Clin Pharmacokinet 51,  Study of genetic polymorphism  Yuichi Sugiyama’s group at the University of Tokyo  Potential to compare healthy metabolism/PK with disease state  Microdosing offers a safe way of studying the effects of, for example, renal or hepatic impairment  Microdosing being applied in late-stage drug development 29

 Microdosing has been around for nearly a decade  Initially it was directed towards early assessment of plasma- based PK  Microdose data provide an early indication of a drug’s plasma PK as well as site of action  It has now evolved into a variety of directions, from investigations into drug-drug interactions, metabolism and study of PK in the pathological condition  Microdose data are preliminary and have to be considered in the context of the questions that need to be addressed and alternative methodologies available Conclusions 30 Microdosing is a tool in the translational science toolbox

 Professor Malcolm Rowland  Professor Yuichi Sugiyama  Xceleron Scientists  Sponsors of the CREAM Trial and EUMAPP Acknowledgements Questions gratefully received Graham Lappin Xceleron Inc 31