INTEGRATED IN VITRO AND IN VIVO APPROACHES TO DRUG METABOLISM INVESTIGATION - A STUDY CASE

Massimiliano Fonsi

Meet the Experts Transporter Conference Seoul - November 14 2019

EVERY STEP OF THE WAY

Introduction

An ideal in vitro DMPK model should include all the pathways that are relevant for the PK in vivo.Quantitative scaling to whole organ/body is essential for the prediction of exposure of parentdrug and its metabolites in vivo.

The liver is generally considered the organ most involved in metabolic transformations and invitro screening systems are based essentially on liver tissue fractions. More frequently,metabolism in the gut is also considered. Conversely, metabolism in other organs is rarely takeninto consideration.

The case study presented today will offer an example of

1) The experimental evidence that should trigger the investigation of extrahepatic metabolism2) The differences that may exist between preclinical species and humans*3) The impact of extrahepatic metabolism on screening cascades and decisions process in a

drug discovery project.

*Extrahepatic metabolism may complicate the IVIVC in rats.Fonsi M - Drug Metab Lett. 2014;8(1):51-66.

2

Discovery of MK-4827 (Niraparib)

Poly(ADP-ribose)polymerase (PARP) as a context specific anti-cancer target

3

• Most anticancer drugs in use today have a low therapeutic index

Can we selectively kill tumor cells?The Synthetic Lethality approach

• Two strategies to improve high therapeutic index

Target-driven - Target only in cancer cells Context-driven

Inhibitor Inhibitor

Survival Death

4

PARPi as context-specific anti-cancer agents

PARP recruited

and Parylation

Cell Survival

DNA single strand break

PARP1

PARP1X

PARP

inhibition

Unrepaired SSB

collapses to DSB

Repaired by

homologous

recombination (HR)BRCA2

BRCA1

Cell

Death

BRCA

deficient cells

PARP1

Cell Survival

PARP inhibition selectively kills

BRCA deficient cancer cells

Recruitment of

BER machinery

5

PARPi’s in BRCA-deficient cells

Selectively kill BRCA-deficient cells in vitro

L-001758092Log [M]

-10 -9 -8 -7 -6 -5

%C

on

t. G

row

th

0

20

40

60

80

100

120

HeLa wtHeLa shBRCA1

CC50= 10uM

CC50= 7nM

Prevents tumor formation in mice when implanted

with BRCA -/- cells but not WT cells

0

20

40

60

80

0 15 30 45 60 75 90 105

Days

Fre

qu

en

cy o

f

Tu

mo

r F

orm

ati

on

(%

)

BRCA2- / PARPi

BRCA2- / Veh

WT / Veh

WT / PARPi

Farmer, Nature 2005

PARP inhibition selectively blocks

growth of BRCA2-deficient tumors

6

First Generation of PARP inhibitors

• First PARPi were based on the cleavage product -nicotinamide

N

O

N

H

H

HO

H O

N

OH

HO

N

O

NH2

SynEnergetically favoured

but not active

AntiActive conformation

N

NH2

O

NH

O

N

R

InternalH-Bond

NH

O

NH

O

NN

N

N

R

H

R

Place amide in heterocyclic ring

Constrain amidein fixed conformation

Indazole series

7

Initial profiling of the prototype PARP inh: early detection of series releted liabilities

N

N

O NH2

R

Where R= Ph

Permeability (Papp) = 5 (cm*10E-6/s)) (LLC-PK1);

ER (BA/AB in LLC-PK1 humMDR1) = 12

ER (BA/AB in LLC-PK1 mouseMdr1) = 17

Enzyme Inhibition

CYP450 (IC50)

3A4, 2D6, 2C9, 2C19, 1A2 >50 µM

Enzyme Induction

hPXR 74% rifampicin effect @ 25 µM (weak inducer)

No CYP3A4 induction in HH @ 30 µM

Off-Target

hERG 9.7 µM

Nav1.5 (0.2 + 3 Hz) 19µM

Solubility 0.02 mg/mL (water);

LogP = 2.4 – TPSA=61

8

NMe2

N

N

NMe

NHMe

NHiPr

NHMe

NHMe

HN

NH

N° R

PARP-1

IC50/nM

PARylation

EC50/nM

rat liver micr.

CLint

((µl/min)/mgP)

rat CL

((ml/min)/kg)

7 3.7 110 177 >250

8 1.9 180 29 107

9 31 1100 34 8

10 3.8 68 28 220

11 6.7 700 18 131

12 3.7 58 9 58

13 16 110 11 30

14 3.1 31 22 ND

15 3.2 24 11 47

N

N

O NH2

R

Rat liver microsomes

R = 0.09

-1

-0.5

0

0.5

1

1.5

2

2.5

3

0.5 1 1.5 2 2.5 3

log in vivo CL [ml/min]

log

liv

er

uS

CL

int

[ul/

(min

*mg

Pro

t)]

rat liver uS

No better correlation using Hepatocyte

Cpds were stable in rat plasma

No improvements correcting CLp by B-to-P ratio or by renal CL

9

Investigating mechanism of extrahepatic CL in vivo of cpd 10…

NN

NH

O NH2

NN

H

O NH2

O

NN

OH

O NH2

O

NN

O-glucuronide

O NH2

O

parent

19

19

20

Urine 0-6h

Bile 0-6h

10

18

19

20

NN

NH

O NH2

NN

H

O NH2

O

NN

OH

O NH2

O

NN

O-glucuronide

O NH2

O

parent

19

19

20

parent

19

19

20

parent

19

19

20

Urine 0-6h

Bile 0-6h

10

18

19

20

IV (treatment) Dose: 3 mg/kg

Formulation: DMSO/PEG400/Water (20%/60%/20%)

Mean std. Dev.

AUC(0-) µM·hr 0,83 0,08

Cl mL/min/kg 217 20

t½ hr 1,4 0,2

Vdss L/kg 22,1 0,8

PO (treatment) Dose: 3 mg/kg

Formulation: PEG 400 (SUSPENSION)

Mean std. Dev.

Cmax µM 0,07 0,04

Tmax hr 2,2 2,5

AUC(0-) µM·hr 0,37 0,14

t½ hr 5,4 3,2

F(0-∞) % 45 17,0

% of dosed

radioactivity

% of the dose recovery

transformed into 19 or

20

dose 100%

urine 0-6h 20% 19%

urine6-24h 21% 19%

bile 0-6h 30% 16%

bile 6-24h 10% 6%

recovery 81% 59,4%

N

N

NH

O NH2

T

10

10

Trying to establish an in vitro model of extrahepatic metabolism

rat CYP phenotyping for cpd 10

0%

20%

40%

60%

80%

100%

120%

0 10 20 30 40 50 60

min

% p

are

nt

1A1

1A2

2A1

2A2

2B1

2C6

2C11

3A2

2C12

2C13

2D1

2D2

3A1

reductase

Metabolic stability in liver, lung, kidney, intestine microsomes and heart homogenate

in presence of NADPH

0

20

40

60

80

100

120

140

0 20 40 60 80 100

time (min)

Perc

ent

of

initia

lcpd 10

Liver nadph

Liver no cof

Lung nadph

Lung no cof

Kidney nadph

Kidney no cof

Intestine nadph

Intestine no cof

Heart nadph

Heart no cof

rat rCYP1A1

R = 0.75

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

0.5 1 1.5 2 2.5 3

log in vivo CL [ml/min]

rCY

P1A

1 lo

g C

Lin

t [u

l/(m

in*n

mo

lPro

t)]

rCYP1A1

rat rCYP1A1

R = 0.75

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

0.5 1 1.5 2 2.5 3

log in vivo CL [ml/min]

rCY

P1A

1 lo

g C

Lin

t [u

l/(m

in*n

mo

lPro

t)]

rCYP1A1

11

0.01

0.1

1

10

0 10 20 30 40 50 60 70

pla

sm

a c

on

cen

trati

on

(µ

M)

Time (min)

Portal vein

Sampling site = jugular vein

Continuous infusion via PORTAL VEIN (Rat SD)

Mechanistic in vivo PK to confirm hepatic and liver organ extraction

12

0.01

0.10

1.00

10.00

0 10 20 30 40 50 60 70

pla

sm

a c

on

cen

trati

on

(µ

M)

Time (min)

Femoral vein

Portal vein

Sampling site = jugular vein

Continuous infusion via FEMORAL VEIN (Rat SD)

13

0.01

0.10

1.00

10.00

0 10 20 30 40 50 60 70

pla

sm

a c

on

cen

trati

on

(µ

M)

Time (min)

Intra-arterial

Femoral vein

Portal vein

Sampling site = jugular vein

Cssfemoral µM 0.52±0.06

Clp* ml/min/kg 140±16

Cssportal µM 0.21±0.06

E%hepatic % 59±19

Cssarterial µM 2.0±0.5

E%lung % 78±23

Continuous infusion via CAROTID ARTERY (Rat SD)

14

SPR in vitro & in vivo

NHMeNHMe NHMe

N

N

O NH2

R

CLint

rCYP1a1 6000 2200 400 300 <100(uL/min/nmol)

CLb

(mL/min/kg) 220 58 30 46 24

NH

(S)

NH

(R)

MK-4827 (Niraparib)

PARylation EC90 (nM) 30 280

15J Med Chem. 2009 Nov 26;52(22):7170-85.

L-001946812-001G005 plasma concentration of Rats receiving

an i.v. infusion via Femoral or Portal vein at 10 g/min for 120 minutes

Infusion via Femoral vein Infusion via Portal veinTime Average Std. Dev. Time Average Std. Dev.

(min) Rat 1 Rat 2 Rat 3 (h) Rat 4 Rat 5 Rat 6

15 0.80 0.85 0.67 0.78 0.09 15 0.51 0.41 0.59 0.50 0.09

30 1.16 1.17 0.79 1.04 0.22 30 0.84 0.78 0.97 0.86 0.09

45 1.36 1.36 0.90 1.21 0.26 45 1.02 0.87 1.15 1.01 0.14

60 1.53 1.52 1.30 1.45 0.13 60 1.61 0.99 1.48 1.36 0.33

90 1.89 2.00 1.69 1.86 0.15 90 1.82 1.47 1.79 1.69 0.19

120 2.29 2.22 1.64 2.05 0.36 120 1.85 1.60 2.15 1.87 0.28

AUC ( M * h) 174.8 177.9 140.5 164.4 20.8 152.9 121.7 159.2 144.6 20.1

Css (µM) 1.8 1.6

Average 60 and 120 min

Project Mol. Wt. Analytical Method Analytical Protocol

PARP 320.397 LC/MS/MS ACN PPT

Date dosed Notebook # LOQ (uM) Analysis date Notebook #

4/1/2007 227351 0.008 11/1/2007 227369

Comments Dosing solution:Saline

Concentration (µM)

Cl (ml/min/kg) = 58E% = 10

Concentration (µM)

0.00

0.00

0.01

0.10

1.00

10.00

0 20 40 60 80 100 120

Time (min)

Co

nc (

uM

)

Femoral vein

Portal vein

N

N

NH2

O

N

H2+

Chiral

Cl

0.10

0.60

1.10

1.60

2.10

2.60

3.10

3.60

0 20 40 60 80 100 120

Time (min)

Co

nc (

uM

)

Femoral vein

Portal vein

N

N

NH2

O

N

H2+

Chiral

Cl

MK-4827 IVIVC: Rat liver data

Hep Intrinsic Clearance

(l/min/million cells)

Rat Hum

2.2 1.6

1 Rat QH = 68 ((ml/min)/kg) - Hum QH = 20 ((ml/min)/kg).

Ring B.J.et al. Pharmaceutical Sciences, 2011 DOI 10.1002/jps

Measured hepatic CL (in vivo)

ERH = 10% CLH= ERH * QH = 6.8 ((ml/min)/kg)1

Predicted Rat hepatic CL1

ERH = 11%

CLH= 7.1 (ml/min)/kg

Predicted hum hepatic CL1

ERH = 14%

CLH= 2.8 (ml/min)/kg

16

TOTAL blood CL = 46 mL/min/Kgbw – F% = 92%

17

Mechanism of Clearance in rats

TOTAL blood CL = 46 mL/min/Kgbw – F% = 92%

• 35% urinary excretion

=> CLR= 0.35*CLB = 0.35*46= 16.1 ml/min/kg ; > GFRrat (5-6 ml/min/kg)-Rat Kidney BF (40-52 ml/min/kg)

• 2 % biliary excretion

• 10 % hepatic extraction (determined by portal/femoral infusion)

• ~ 50 % extrahepatic metabolism

17

Transporters Data

Compound PappB-A/PappA-B

LLC-PK1

Papp (cm*10E-6/s)

Human

MDR1

Mouse

Mdr1a

Rat Mdr1a Beagle

MDR1

Monkey

MDR1

MK-4827

(5 M)

3 12 22 17 10 ND

Verapamil

(1 M)

1 6 12 7 4 ND

MK-4827 is a Pgp substrate

Solubility >20 mg/mL

24

Turnover in lung and kidney preparations

MK-4827 in NADPH-supplemented preparations:

• rat lung microsomes (40 % turnover in 90 min)

• rat kidney S9 fraction (50% turnover in 90 min)

• human lung microsomes, from both non-smokers and smokers (stable)

• human kidney S9 fraction (stable).

• These results suggest low contribution of extra-hepatic metabolism in humans.

• Predicted low hepatic CL in human (IVIVE using HH)

• Potential renal CL in human due to Low PPB – Mdr1 active efflux

• Expected good F%

18

20

Proposed Metabolite Scheme (Rat and Human)

NH

NH2O

N

N

m/z 321

NH

NH2O

N

N

-H2+O

m/z 335

NH

NH2O

N

N

O

m/z 337

NH

NH2O

N

N

-H2+O2O

m/z 367

NH

NH2O

N

N

O

NH

NH2O

N

N m/z 351

-H2+OO NH

NH2O

N

N

O2

m/z 353

MK-4827

RlungMHLM,RLMRat bileRat urine

RLM,HLMrCYP1A1 H,RRat urine HLM, RLM

Rat bileRat urine

Rat urine

Rat bileRat urine

Rat urineRat bile

m/z 337

NH

OHO

N

N

m/z 322

HLM, RLMHH, RHRat urine, bileTraces

NH

NH2O

N

N

m/z 479 Gluc

M1

M2

M3

M4

M5

M6

M7

M8

Rat bile

Traces

Activity in MDA-MB-436 (BRCA1-) xenografts

0

200

400

600

800

1000

1200

1 5 8 14 21 29 33

Tu

mo

r vo

lum

e (

mm

3)

DAY

Vehicle QD

Vehicle BID

MK4827 100 mpk QD

MK4827 50 mpk BID

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Tumor Weight (g)

0

20

40

60

80

100

120

1 5 8 14 21 29 33

% B

od

y W

eig

ht

DAY

P < 0.005

P < 0.005

P < 0.005

P < 0.005

Administered orally

Approved as PCC April 2007 (MK-4827)20

22

Species Dose

(mg/kg)

Clb

(mL/(min*kg))

Vd

(L/kg)

T1/2

(hr)

AUC

(M*hr)0-x

Rat 3 46±3 22.8±8.7 5.1±2.0 2.9±0.1

Dog 1 89±6 29.3±5.2 5.3±0.3 0.6±0.0

Monkey 1 17.5±6 17.1±2.5 6.8±1.0 1.5±0.2

± ± ± ±

IV (male, n=3, saline)

Species Dose

(mg/kg)

Cmax

(M)

Tmax

(hr)

AUC

(M*hr)0-x

%F

Rat 4 0.3±0.1 4.0±1.0 3.7±1.8 92±29

Dog 2 0.2±0.0 0.9±0.3 0.8±0.1 80±14

Monkey 2 0.3±0.1 1.3±0.6 1.9±0.5 70±17

± ± ± ±

P.O. (male, n=3, water)

MK-4827 (Niraparib) preclinical profiling - PK

Dose proportional PK in rat – MK-482721

23

Proposed Metabolite Scheme (Dog)

NH

NH2O

N

N

m/z 321

NH

OHO

N

N

m/z 322

MK-4827 L-00958601 hepatocytes microsomes urine plasma

PLASMA Concentrations (uM) of L-001946812 in 1 mg/kg IV Dosed

DOG BEAGLE

0.001

0.01

0.1

1

0 5 10 15 20 25 30

Time (Hour)

(Log)

Concentr

atio

n (

uM

)

Mean plasma conc of L-001946812

Mean plasma conc of L-001958601

M7

22

24

Rat/Dog Excretion Study 3 mg/kg iv (n=3)

Percent of Dose Recovered (0-48h)

urine bile (0-24h) total

Total 48±1 32 ±4 80±4

Parent 35±2 2±2 37±3

Water 2±1 2±1

Data obtained using MK4827 labeled with tritium in position 3 of the indazole ring.

Low tritium loss was detected both in vivo and in vitro

In dogs only 2 % of parent excreted in urine

23

Hum CL – t1/2 prediction for MK4827

• Total hum CL = CLmetab liver + CLR + CLmetab extrahep

• CLmetab liver : IVIV scaling from HH = 2.8 ml/min/kg (range 1.8-3.8)

• CLR = CLR-rat * (fuhum / furat )* KBFhum / KBFrat * ERhum/ERrat= 4.5 ml/min/kg

(range 3.5-7)

• CLmetab extrahep: expected to be negligible in human

• Vdss = Vdss hum = Vd (animal)*fu(hum)/fu(animal) = 26 (L/kg) range (17-37)

25

Vd (L/kg)17 26 37

total min avrg maxCL 5.5 min 28 43 61

(mL/min/kg) 8.3 avrg 21 36 4611 max 16 25 36

Values used for rat, and human KBF were 52 (Hsu et al., 1975) and 16 ml/min/kg (Wolf et al., 1993)

Predicted half-life (hr)

25

Human PK data (Phase II)

http://meetinglibrary.asco.org/content/62393?format=posterImg

Niraparib was approved in 2017

26

Case studies in drug discovery have been reported showing relevant lungs and generalized extrahepatic metabolism in rats. The large extrahepatic contribution appeared to be rat specific with no analogy to human metabolism. Rat CYP1a1 was identified as the major enzyme responsible

The extraction fraction (CL/Q) in the lung is in general small, as a result of its modestmetabolic activity but due to the fact that the lung blood flow is equivalent to the totalcardiac output, in some cases, drug metabolism in the lungs can be substantial.

In general, xenobiotic metabolizing enzymes (XMEs) are expressed in lower levels in the extrahepatic tissues than in the liver, making the former less relevant for the clearance of xenobiotics. Local metabolism, however, may lead to tissue-specific adverse responses, e.g. organ toxicities, allergies or cancer.*

Conclusion

Fonsi M - Drug Metab Lett. 2014;8(1):51-66.* U. Gundert-Remy et al. Drug Metab Rev, Early Online: 1–34

27

Thank you very much!