HµREL Symposium (2015 ISSX Meeting)

Comparison of HµREL Co-culture Hepatocyte Model to Conventional In Vitro Models for Intrinsic Clearance and Metabolite ID

J. Matthew Hutzler, Ph.D.Director, In Vitro Metabolism

2

In Vitro Metabolism and ScalingPredicting Clearance

Common methodology to evaluate metabolic stability is to incubate test compound (0.1 to 1 µM, or <<Km) with liver microsomes or hepatocytes and measure rate of substrate depletion (half-life)

0 15 30 45 600

1

2

3

4

5

Compound 1

Compound 2

No NADPH

Time (min)

LN

% R

emai

nin

g

Clint = 0.693 x 25.7 g liver x mL inc x 45 mg mic protein = mL/min/kg t1/2 (min) kg body Wt mg mic protein g Liver

CLh = Qh x fu x CLint

Qh + (fu*CLint)

Slowly metabolized (i.e. low-turnover) drug molecules are becoming increasingly common For low-turnover compounds, traditional incubation systems (e.g. microsomes and hepatocyte

suspensions) are inadequate for a reliable clearance projection due to incubation time limitations (~1-2 hr for microsomes and 4 hr for hepatocytes)

Clearance projectionWell-stirred model

Hutzler et al. (2015) Drug Metab Dispos. 43(12)

3

Cryopreserved Human Hepatocyte SuspensionLiterature Data Demonstrating Rapid Loss in Activity

Smith et al. (2012) J Pharm Sci. 101(10), 3989-4002

Effect of time (in humidified incubator at 37 C) on activities in cryopreserved hepatocyte suspensions

Mean IT50 values (time where 50% loss in activity was observed)• CYP1A2: 2.69 hrs• CYP2C9: 4.47 hrs• CYP2D6: 3.03 hrs• CYP3A: 1.62 hrs• UGT: 1.39 hrs

Rapid loss of activity that is enzyme-dependent

Incubation time limit: ~4 hrs

4

In Vitro Assay Option for Longer Duration IncubationsHµREL® Hepatic Co-culture Models

HµREL® - a hepatic co-culture platform (with stromal cells)

Robust enzyme activity for days to weeks

Plates arrive ready to use same day

Multiple plate formats available (12, 24, 48, 96 well)

Human, Rat, Dog, Monkey available on same plate for cross-species comparisons

Can use any plateable lot of hepatocytes

Q2 Solutions has conducted pilot studies with HµREL® co-culture system in effort to evaluate and potentially utilize for clients with low turnover compounds needing either clearance projection, or metabolite profile information

Leslie Benet, Presentation at The Boston Society, July 2013

5

Low Clearance In Vitro Assays: A Comparison StudyStudy Design Summary

Stromal cell only plate included as a control for HµREL system study

Condition Suspension (4 hr) Plated Monolayer (24 hr) HµREL® (72 hr)

Hepatocyte Lot LiverPoolTM 5-Donor Mixed Gender Cryoplatable Lot YMD (BioreclamationIVT)

Thawing Media CHRM CHRM HµREL thaw

Plating Media - Hu Hep Plating Media HµREL plate

Incubation Media HMM HMM HµREL Dosing Media

Seeding density (cells/well) 50,000 45,000 30,000

Inc Volume 200 µL 100 µL 100 µL

Substrate Concentration 1 µM 1 µM 1 µM

Plate Design 96 well 96 well 96 well

Shaker speed 600 rpm 120 rpm 120 rpm

Time Points (6) 0, 15 min, 0.5, 1, 2, 4 hr

0, 4, 6, 8, 18, 24 hr 0.25, 3, 6, 24, 48, 72 hr

General procedures Thaw cells, incubate in suspension for total of

4 hrs

Plate at 8am, dose at noon (t0), sample at 4 pm, 6 pm, 8 pm, 6 am, and noon next

day

Receive plates on a Tuesday morning,

dose in afternoon, 72 hrs is Friday afternoon

6

Low Clearance In Vitro Assays: A Comparison StudySelected Substrates

Compound Enzymes Involved in Metabolism

± Warfarin CYP2C9, CYP3A, CYP1A2

Timolol CYP2D6

Diazepam CYP3A, CYP2C19

Tolbutamide CYP2C9

Theophylline CYP1A2

Metoprolol CYP2D6

Efavirenz CYP2B6

Verapamil CYP2D6, CYP3A, UGT

Diclofenac CYP2C9, UGT

Alprazolam CYP3A

Glimepiride CYP2C9

Meloxicam CYP2C9

Prednisolone CYP3A

13 substrates selected based on range of reported clearance values, as well as diversity in cytochrome P450 metabolic clearance mechanisms

Minimal renal clearance

Clearance rates estimated by substrate depletion methodology

Scale-up to intrinsic clearance

7

1) kdep = slope of time (min) vs. Ln% Remaining;

2) In vitro half-life:

3) In Vitro Intrinsic Clearance:

4) Intrinsic Clearance (Clint):

5) Clint in vivo back-calculated from Clin vivo using the well-stirred model

Calculations of Intrinsic Clearance

𝑡1/2 = 𝐿𝑛(2)𝐾𝑑𝑒𝑝

𝐿𝑛(2)𝑡1/2 𝑥 𝐼𝑛𝑐 𝑉𝑜𝑙 (µ𝐿)𝐶𝑒𝑙𝑙𝑠 𝑥 106 = µ𝐿min ∗106 𝐶𝑒𝑙𝑙𝑠

𝐿𝑛(2)𝑡1/2 𝑥 𝐼𝑛𝑐 𝑉𝑜𝑙 (𝑚𝐿)𝐶𝑒𝑙𝑙𝑠 𝑥 106 𝑥 120𝑥106𝐻𝑒𝑝𝑎𝑡𝑜𝑐𝑦𝑡𝑒𝑠𝑔 𝐿𝑖𝑣𝑒𝑟 𝑥 25.7 𝑔 𝐿𝑖𝑣𝑒𝑟𝐾𝑔 𝐵𝑊𝑡 = 𝑚𝐿min ∗𝐾𝑔

8

Representative Metabolic Stability Figures

0 50 100 150 200 2500

1

2

3

4

5

Warfarin

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Timolol

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Theophylline

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Warfarin

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Timolol

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Theophylline

Time (min)

LN

% R

emai

nin

g

0 1000 2000 3000 40000

1

2

3

4

5

Warfarin

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Timolol

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Theophylline

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 50 100 150 200 2500

1

2

3

4

5

Diazepam

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Diazepam

Time (min)

LN

% R

emai

nin

g

0 1000 2000 3000 40000

1

2

3

4

5

Diazepam

Time (min)

LN

% R

emai

nin

g

HrelStromal

Su

spen

sio

n P

late

d H

µR

EL

®

NC = no clearance

NC NC NC

NC NC NC NC

NC45% dep

70% dep77% dep

*

*internal standard issue

9

Representative Metabolic Stability Figures S

usp

ensi

on

Pla

ted

Hµ

RE

L®

0 50 100 150 200 2500

1

2

3

4

5

Tolbutamide

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Metoprolol

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Efavirenz

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Diclofenac

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Tolbutamide

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Metoprolol

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Efavirenz

Time (min)

LN

% R

emai

nin

g

0 500 1000 15000

1

2

3

4

5

Diclofenac

Time (min)

LN

% R

emai

nin

g

0 1000 2000 3000 40000

1

2

3

4

5

Tolbutamide

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Metoprolol

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Efavirenz

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Diclofenac

Time (min)

LN

% R

emai

nin

g

HrelStromal

NC

NC = no clearance

88% dep

96% dep91% dep

10

Intrinsic Clearance Comparison(mL/min/kg)

Hepatocyte suspension assay yields higher turnover than either plated format for moderate to high clearance compounds (consistent with literature) Clearance rate in HµREL co-culture system 2- to 7-fold higher than hepatocyte monoculture format, with better ability to measure clearance rate (11/13 compounds, or 85%)

11

In Vitro-In Vivo Correlation Analysis

Suspended hepatocyte data correlates best overall for moderate/high clearance, but incubation time limits measuring clearance for low-turnover compounds

Plated monoculture performs poorly (5/9 non-detects), and 24 hour incubation may not be long enough for a low-turnover compound

HµREL co-culture trends toward under-prediction for moderate/high clearance compounds, but preferred over monoculture system

Additional studies with more low-turnover compounds under way

prednisolone

warfarin

tolbutamide

Drop by Poster P10 to discuss more…

12

Guidance Expectations around Metabolite Profiling (FDA/EMA/Japan): Are there “major” circulating human metabolites, and are they covered in the toxicity species - MIST Are there “major” circulating human metabolites that may elicit clinically relevant drug interactions –

DDI What are the primary metabolic clearance pathways, including the enzymes responsible – DDI

Prior to Phase I dosing, in vitro systems are relied upon to predict human metaboliteso To support selection of toxicity specieso To predict the potential for victim-based DDIs and/or PK variability in the clinico For moderate/high clearance drugs, traditional in vitro systems are used for these assessments

For low clearance drugs, HLMs or suspended hepatocytes are often inadequate at investigating metabolic pathways

Metabolite Identification

Study Design

4 hour suspension hepatocyte incubation 24-well plate design (HµREL), with ~188K hepatocytes/well 4, 24, 72, and 168 hr incubations (no media change) 10 µM meloxicam, timolol, and XK-469 Analysis using high-resolution mass spectrometry (Thermo Orbi-TRAP) Comparison of metabolite profiles

13

Meloxicam Metabolism

10 15 20 25 30 35Time (min)

Melox-OH

Melox-COOHHµREL – 4 hr

HµREL – 24 hr

HµREL – 72 hr

HµREL – 168 hr

Standard Hepatocyte4 hr Incubation

Meloxicam

Melox + 4H + O +Melox + 2 H2O

Melox-OH

Suspended hepatocytes: Observed only trace levels of

+O (Melox-OH)

HµREL Co-culture: Observed progression from

+O (Melox-OH) to subsequent acid (Melox-COOH) over time

Robust metabolism to yield multiple metabolites

Comparison to in vivo data: Based on 14C-human:

o Melox-OH ~9% doseo Melox-COOH ~60% doseTurck et. al. 1996

Conclusions: Melox-COOH and Melox-OH

readily observed\ Better correlation with in

vivo met ID data +4H+O and +2H2O peak

significant in HµREL, in vivo relevance unknown

SN

OH

O O

NH

O

S

N

Meloxicam

SN

OH

O O

NH

O

S

NOH

SN

OH

O O

NH

O

S

NOH

O

Melox-OH Melox-COOH

14

Timolol Metabolism

10 15 20 25 30Time (min)

Timolol

Timolol + Gluc

+ O

+ 2O

+ OMorpholine ring open+2O

- C2H2

HµREL – 4 hr

HµREL – 24 hr

HµREL – 72 hr

HµREL – 168 hr

Standard Hepatocyte4 hr Incubation

+ O

Suspended hepatocytes: Observed only trace levels of

metabolites, largest +O

HµREL Co-culture: Seeing progression from +O

to morpholine ring opening (+2O) to loss of ethyl (-C2H2)

Seeing robust metabolism to yield multiple metabolites

Comparison to in vivo data: Based on 14C-human:

o Morpholine ring open (+2O) ~22% dose

o Loss of ethyl bridge from morpholine ~7% dose

Tocco et. al. 1980

Conclusions: Morpholine ring opening with

subsequent loss of ethyl bridge readily observed

Intriguing that +gluc was not observed in suspensions but significant in HµREL

O

N

N

SN

O

HNHOTimolol

N

N

SN

O

HNHO

HO

O

OH

Morpholine ring open (+2O)

HN

N

SN

O

HNHO

OH

Loss of ethylbridge

Loss of ethyl bridge

15

XK-469 Metabolism

Suspended hepatocytes: Observed only trace levels

metabolites, largest +O

HµREL Co-culture: Seeing +O associated with

aldehyde oxidase as largest peak

Seeing robust metabolism to an array of non-P450 metabolites

Comparison to in vivo data: Based on human data:

o AO metabolite accounts for 54% of total urinary excretion, and is most predominant metabolite

Anderson et. al. 2005

Conclusions: Clearly suggests non-P450

pathways such as AO are viable for 7 days

A good example of the need for a reliable in vitro system for prediction of metabolic pathways in human10 15 20 25 30 35

Time (min)

HµREL – 24 hr

HµREL – 72 hr

HµREL – 168 hr

HµREL – 4 hr

XK469

N

N

O

OOH

O

Cl N

N

O

OOH

O

Cl

XK469

OH

AO-mediated hydroxylation

+ O (AO) + Glycine

+ Taurine

+ AcetylCys – 2H

+ Cys – 2H+ Cys – 2H

Standard Hepatocyte4 hr Incubation

16

Acknowledgements

Mark VandenBranden Todd Hieronymus Andy Staton Barbara Ring

Clearance Studies

Metabolite ID Studies

Taysir Chamem Todd Hieronymus David Heim Rich Burton Shelby Anderson Xiusheng Miao

HµREL

Eric Novik Matt Shipton

17

Back-Up Slides

18

Company Confidential

Time (min.) % B0 52 542 X Xmeloxicam = 60

42.1 95 Xtimolol = 2047 95 XXK469 = 70

47.1 560 5

• Mass Spectrometry› System: Thermo LTQ Orbitrap XL

› Ionization: ESI+; Source: 5.0 kV; Capillary: 350゜ C

› Fragmentation Method: HCD

› Collision Energy: 25 (meloxicam), 50 (timolol), 40 (XK469)

• Chromatography› System: Shimadzu Nexera X2 UHPLC

› Column: Supelco Ascentis Express C18, 2.7 µm, 2.1×150mm; 25゜ C

› Mobile Phase: A: 10 mM ammonium formate + 0.1% formic acidB: acetonitrile

› Flow Rate: 0.15 mL/min.

› Gradient:

Instrumentation/Method Utilized

19

Metabolic Stability Data Figures S

usp

ensi

on

Hµ

rel®

0 50 100 150 200 2500

1

2

3

4

5

Glimepiride

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Prednisolone

Time (min)

LN

% R

emai

nin

g

0 50 100 150 200 2500

1

2

3

4

5

Meloxicam

Time (min)

LN

% R

emai

nin

g

0 1000 2000 3000 40000

1

2

3

4

5

Alprazolam

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 50 100 150 200 2500

1

2

3

4

5

Alprazolam

Time (min)

LN

% R

emai

nin

g

0 1000 2000 3000 40000

1

2

3

4

5

Glimepiride

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Prednisolone

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Meloxicam

Time (min)

LN

% R

emai

nin

g

HrelStromal

NC NC

NC = no clearance

NC51% dep 53% dep

Only theophylline and alprazolam showed no clearance in Hurel® system

20

0 100 200 3000

1

2

3

4

5

Timolol_Suspension

Time (min)

LN

% R

emai

nin

gIn Vitro Metabolism and ScalingExtrapolation Guidelines

• A cut-off for half-life extrapolation using in vitro data needs to be established

• To confidently extrapolate an in vitro half-life, at least 20% depletion of substrate needs to be observed to be interpreted as metabolism (i.e. beyond biological/analytical noise)

• 20% depletion at t240 min roughly equates to a half-life ~3x the incubation time (~720 min), but all data points need to be considered

• An in vitro system enabling a longer incubation time would be beneficial to improve our confidence in prediction

t1/2 = 630 min(2.6x > inc time)

104 98 91 80 • Assuming reasonable extrapolation of half-life to ~3x beyond length of assay (extrapolate t1/2 to 720 min): • Clint limit: ≤11.9 mL/min/kg• Clh limit: ≤7.5 mL/min/kg (36% LBF)

• Human liver blood flow (LBF): 20.7 mL/min/kg

21

Low Clearance In Vitro AssaysOptions and Pros/Cons

Relay Method Using Cryopreserved Hepatocytes – Relay every 4 hours

Pros Cons

Utilizes standard max 4 hr incubation time (literature support) Uses 4-5x the amount of hepatocytes (~$3000 in hepatocytes)

Can use any lot of pooled cryopreserved hepatocytes Labor intensive (4-5 days), depending on number of relays

Good correlation with human in vivo data (demonstrated IVIVC)

Have to account for lost compound during relays (calculation)

IVIVC also demonstrated for pre-clinical species (rat/dog) Dilution factor for each relay

Number or relays flexible (max of 5) Challenge to reproduce results (personal communication)

* Relay method also shown to be suitable to generate metabolites for met ID

Thaw at T=0 hrs Thaw at T=4 hrs Thaw at T=16 hrsThaw at T=12 hrsThaw at T=8 hrs

Di et al. (2012) Drug Metab Dispos. 40, 1860-1865Di et al. (2013) Drug Metab Dispos. 41, 2018-2023*Ballard et al. (2014) Drug Metab Dispos. 42, 899-902

*assuming ~$600 per vial

22

Low Clearance In Vitro AssaysOptions and Pros/Cons

Plateable Cryopreserved Human Hepatocytes – single donor or plateable pool

Pros Cons

Enzyme activity more stable compared to suspension Enzyme activity decreases over first 24 hrs (~50%)

Practical solution – plate cells and dose 4 hrs later Not all lots of hepatocytes are plateable

Some plateable pools now available (5 or 10 donors) There is question about the plating efficiency of multiple donors; single donor, enzyme variability a concern (no PMs)

24 hrs incubation limit may not be long enough for low CL

• Effects on specific activities in plated cryopreserved hepatocytes (4 preparations)• IT50 estimates for CYP1A2 and CYP3A: 21.3 ± 2.1 and 28.8 ± 20.4 hrs

Smith et al. (2012) J Pharm Sci. 101(10), 3989-4002

* Hepatocytes were plated and allowed to attach for 4 hours prior to initiation of study

23

Metabolic Stability Data Figures

0 50 100 150 200 2500

1

2

3

4

5

Verapamil

Time (min)

LN

% R

emai

nin

g

Su

spen

sio

n

Pla

ted

Hµ

rel®

0 500 1000 15000

1

2

3

4

5

Verapamil 1

Time (min)

LN

% R

emai

nin

g

0 1000 2000 3000 40000

1

2

3

4

5

Verapamil 1

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

Verapamil 2

Time (min)

LN

% R

emai

nin

g

HrelStromal

0 1000 2000 3000 40000

1

2

3

4

5

6

Verapamil 3

Time (min)

LN

% R

emai

nin

g

HrelStromal

Verapamil 1-3 all different plates; verapamil 3 run on different date