approaches to mitigating the bioactivation potential of ......alpidem (anxiolytic) hepatitis (fatal)...

Deepak Dalvie 1

Approaches to Mitigating the Bioactivation Potential of Compounds in Lead Optimization

Deepak Dalvie

Pharmacokinetics, Dynamics and Metabolism Department

Pfizer San Diego

Deepak Dalvie 2

Question

Is covalent binding or GSH Adduct Formation a kill shot for a candidate??

Deepak Dalvie 3

Adverse Drug Reactions (ADR) A Problem??

è  A common cause for drug recalls or black box warnings

è  Unpredicted or Idiosyncratic ADR (IADR) – more problematic Ø  Normally not observed until phase III or post launch Ø  Frequency of occurrence – 1 in 10000 to 1 in 100000 Ø  Responsible for drug withdrawal Ø  Very expensive - law suits etc.

è  Patients have been deprived of several good drugs

Deepak Dalvie 4

è  Not easy to predict IADR l  No animal models l  Lack of specific biomarkers

è  Circumstantial evidence links metabolic activation to IADR

Drug Safety - One of the leading causes for candidate attrition

Drug Reactive

Metabolite Formation

Interaction with Macromolecules

Haptenization Reaction with proteins

ADR Liver toxicity

Agranulocytosis Other

Metabolism

è  Several examples demonstrate that bioactivation liability and IADR are linked l  Kalgutkar AS and Soglia JR (2005). Exp. Opin. Drug Metab. & Toxicol. 1:91-141)

Deepak Dalvie 5

Drugs Associated With IADRs

Drugs Withdrawn Aclcofenac (antiinflammatory) Hepatitis, rash Alpidem (anxiolytic) Hepatitis (fatal) Amodiaquine (antimalarial) Hepatitis, agranulocytosis Amineptine (antidepressant) Hepatitis, cutaneous ADRs Benoxaprofen (antiinflammatory) Hepatitis, cutaneous ADRs Bromfenac (antiinflammatory) Hepatitis (fatal) Carbutamide (antidiabetic) Bone marrow toxicity Ibufenac (antiinflammatory) Hepatitis (fatal) Iproniazid (antidepressant) Hepatitis (fatal) Metiamide (antiulcer) Bone marrow toxicity Nomifensine (antidepressant) Hepatitis (fatal), anaemia Practolol (antiarrhythmic) Severe cutaneous ADRs Remoxipride (antipsychotic) Aplastic anaemia Sudoxicam (antiinflammatory) Hepatitis (fatal) Tienilic Acid (diuretic) Hepatitis (fatal) Tolrestat (antidiabetic) Hepatitis (fatal) Troglitazone (antidiabetic) Hepatitis (fatal) Zomepirac (antiinflammatory) Hepatitis, cutaneous ADRs

Marketed Drugs Abacavir (antiretroviral) Cutaneous ADRs Acetaminophen (analgesic) Hepatitis (fatal) Captopril (antihypertensive) Cutaneous ADRs, agranulocytosis Carbamazepine (anticonvulsant) Hepatitis, agranulocytosis Clozapine (antipsychotic) Agranulocytosis Cyclophosphamide (anticancer) Agranulocytosis, cutaneous ADRs Dapsone (antibacterial) Agranulocytosis, cutaneous ADRs, aplastic anaemia Diclofenac (antiinflammatory) Hepatitis Felbamate (anticonvulsant) Hepatitis (fatal), aplastic anaemia (fatal), severe restriction in use Furosemide (diurectic) Agranulocytosis, cutaneous ADRs, aplastic anaemia Halothane (anesthetic) Hepatitis Imipramine (antidepressant) Hepatitis Indomethacin (antiinflammatory) Hepatitis

Isoniazid (antibacterial) Hepatitis (can be fatal) Phenytoin (anticonvulsant) Agranulocytosis, cutaneous ADRs Procainamide (antiarrhythmic) Hepatitis, agranulocytosis Sulfamethoxazole (antibacterial) Agranulocytosis, aplastic anaemia Terbinafine (antifungal) Hepatitis, cutaneous ADRs Ticlopidine (antithrombotic) Agranulocytosis, aplastic anaemia Tolcapone (antiparkinsons) Hepatitis (fatal) Trazodone (antidepressant) Hepatitis Trimethoprim (antibacterial) Agranulocytosis, aplastic anaemia, cutaneous ADRs Thalidomide (immunomodulator) Teratogenicity Valproic acid (anticonvulsant) Hepatitis (fatal), teratogenicity

Temp. Withdrawn or Withdrawn in other Countries

Aminopyrine (analgesic) Agranulocytosis Nefazodone (antidepressant) Hepatitis (> 200 deaths) Trovan (antibacterial) Hepatitis Zileuton (antiasthma) Hepatitis

For most of these drugs, bioactivation to reactive metabolites has been demonstrated in vitro or in vivo

Kalgutkar AS and Soglia JR (2005). Exp. Opin. Drug Metab. & Toxicol. 1:91-141)

Deepak Dalvie 6

è  Identify toxicophores – “Structural Alerts” early on l  Especially those undergoing ‘metabolic activation’

è  One approach adopted by many pharmaceutical companies l  Eliminate metabolic activation of a candidate

Ø  Screen for ability to form reactive metabolites very early

l  Avoid chemical functionalities that undergo bioactivation Ø  No guarantee that this will make compounds safer Ø  But, preventing reactive metabolite formation ê chances of IADR

è  TALL ORDER BUT AVOIDS RISK! –  Saves $$ –  Saves time and effort

Role of Drug Metabolism in Addressing IADR

Deepak Dalvie 7

Methods to Assess Bioactivation Potential

Covalent Bindingto Proteins

Trapping with nucleophiles

Two Methods Commonly Used

è  Most definitive è  Useful in quantitation of

the reactive metabolite è  Helps to detect all reactive

metabolites è  Limited by availability of

radiolabel

è  Most popular in a discovery setting

è  Nucleophiles used to trap are: l  Glutathione l  N-Acetylcysteine l  Other nucleophiles - cyanide

è  Acts as a surrogate marker of covalent binding

Evans D. C. et. al. Chem. Res. Toxicol. 2004

Deepak Dalvie 8

General Method for Assessing Reactive Intermediates

Drug (10 – 50 µM) + Human Liver Microsomes + GSH (3 mM)

Incubation at 37º C for 1.0 hr

Samples analyzed for GSH adducts using various LC-MS methods

NADPH regenerating system (+/-)

Deepak Dalvie 9

What Does a Positive Signal in These Assays Mean?

Some Known Facts: è  No assay predicts potential of a new drug to cause

IADR

è  Important to Weigh the Benefits versus Risks Ø  Is it a prototype or a back up? Ø  Indication Ø  Is the drug being developed for an unmet medical need?

–  First in class? Ø  Dosing regimen of the drug

–  Acute or chronic? –  Drugs that are used chronically are more prone to IADRs

Deepak Dalvie 10

Relationship Between Dose and Frequency of Idiosyncratic Reactions

è  Dose l  A lower dose reduces the risk

02468

1012

0-50

50-1

00

100-

200

200-

400

400-

800

800-

1600

>160

0

Daily dose (mg)

Freq

uenc

y

Issue: è  Dose is unknown in the early stages

ê Dose ê Lesser reactive metabolite formed ê Lower Risk

Deepak Dalvie 11

Other Factors from a Drug Metabolism Perspective

è  Contribution of pathways that detoxify reactive metabolites

è  Contribution of competing metabolic routes

DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4

DetoxicationDetoxication of of reactive intermediatesreactive intermediates

Competing metabolicCompeting metabolicpathwayspathways

DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4



“After all it is all about body burden of RM”

Deepak Dalvie 12

Case Studies

è  Paroxetine – an antidepressant SSRI è  Raloxifene – A selective estrogen receptor modulator

(SERM) è  A Case Study from Pfizer

Deepak Dalvie 13

Paroxetine Case

è  Paroxetine contains the methylenedioxy group - a toxicophore l  Results in inactivation of CYP2D6

Ø  clinical pharmacokinetic interactions with substrates well established

§Bertelsen KM, Venkatakrishnan K, Von Moltke LL, Obach RS and Greenblatt DJ (2003) Drug Metab. Dispos. 31:289-293

Methylenedioxy Moiety

Carbene

NH

F

O

O

O

Paroxetine

P450 2D6

O

OO

MI Complex withHeme

P450 Inactivation

O O

NN

NN

HOO

HOO

Fe

O O

NN

NN

HOO

HOO

Fe

An Carbene – Iron Complex

Deepak Dalvie 14

Paroxetine Also Undergoes Metabolic Activation

Zhao SX, Dalvie DK, Kelly JM, Soglia JR, Frederick KS, Smith EB, Obach RS and Kalgutkar AS (2007) Chem. Res. Toxicol. 20:1649-1657.

Incubation with HLM in the presence of GSH and NADPH

NH

F

O O

O

OO

O

O

OH

OH

CYP2D6

Catechol o-Benzoquinone

O

OH

OHSG

O

O

OSG

[O]

O

OH

OHSG

GSHGS

bis-Adduct mono-Adduct

[GSH]

Paroxetine

O

O

OSG

GS N

S

O

O

HN

CO2H

SG

[O]

A Novel CyclicAdduct

Deepak Dalvie 15

Incubation of Radiolabeled Paroxetine with HLM


+NAD

PH

+GSH

+NAD

PH

+UDP

GA +N

ADPH

-NADP

HCov

alen

t Bin

ding

(pm

ol b

ound

/mg

prot

ein)

0

10

20

30

40

50

Incubation of radiolabel with HLM/NADPH

[14C] Paroxetine + Human Liver Microsomes +

NADPH

Covalent Binding to Microsomal Proteins

Deepak Dalvie 16

Inference?

è  Paroxetine is prone to bioactivation l  Covalently binds to microsomal protein

è  Would one nominate Paroxetine if it was to be developed

as an antidepressant to date?

è  Paroxetine is a commonly prescribed antidepressant l  IADRs (especially hepatotoxicity) are extremely rare

Deepak Dalvie 17

Effect of GSH on Covalent Binding of Paroxetine

+NAD

PH

+GSH

+NAD

PH

+UDP

GA +N

ADPH

-NADP

HCov

alen

t Bin

ding

(pm

ol b

ound

/mg

prot

ein)

0

10

20

30

40

50Incubation with Human Liver Microsomes


O

O

OO

OH

OHCatechol o-Benzoquinone

[O]

Glutathione Adducts

Covalent Binding

GSH and UDPGA Mops Reactive Intermediate Reduces covalent binding!

Deepak Dalvie 18

+NAD

PH

+GSH

+NAD

PH

+PAP

S+NA

DPH

+SAM

+NAD

PH

+UDP

GA +N

ADPH

-NADP

HCov

alen

t Bin

ding

(pm

ol b

ound

/mg

prot

ein)

0

1

2

3

4

5

6

O

O

OO

OH

OHCatechol o-Benzoquinone

[O]

O

OCH3

OH O

OH

OCH3

+

COVALENTBINDING

Covalent Binding of Paroxetine in Human S9

Sulfate and Glucuronide Conjugates

GSH Conjugates

/COMT S-Adenosyl methionine

Catechol detoxified via methylation O-Methoxy derivatives formed

Reduces covalent binding! This is in addition to GSH conjugation!


Deepak Dalvie 19

Analysis of the Data

è  Paroxetine was bioactivated but: l  o-Quinone is detoxified by Glutathione l  Catechol detoxified by methylation and glucuronidation

è  Additional factors that may contribute to less IADRs with paroxetine l  Low daily dose (20 mg QD)

Ø  Reactive metabolite burden readily handled by endogenous glutathione pool in the mammal

l  CYP2D6 inactivation by paroxetine following chronic dosing may inhibit its own metabolism

Ø  Decreases the body burden of the catechol and o-benzoquinone

DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4



DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4



Drug Catechol

Quinone GSH Conjugates

Methylation Sulfate & Glucuronide

Conjugates

Deepak Dalvie 20

Lesson

è  Thorough knowledge of ALL metabolism pathways is important

è  GSH conjugate detection needs to be put into right context l  GSH is a detoxication pathway! l  Acts as a electrophile mopping agent in the body

è  Early assessment of dose can help

Deepak Dalvie 21

Raloxifene Case

è  A second generation SERM approved for osteoporosis è  Bioactivated by CYP3A4 to quinonoid intermediates è  A mechanism-based inactivator of CYP3A4 è  No IADRs or DDIs reported

SHOOH

O O

N

RaloxifeneSO

O

O O

RaloxifeneDi-quinone methide

SHOOH

O O

SHOOH

O O

SHOOH

O O

GS

SG

SG

[CYP3A4]GSH

GSH Conjugates

Abs

orba

nce uA

U

Abs

orba

nce uA

U G1

100000 150000 200000 250000

M2 M1 Raloxifene

G2

Time (min)

G1

100000 150000 200000 250000

M2 M1 Raloxifene

G2

Time (min)

Incubation with HLM/NADPH and GSH

So would one nominate this compound for further development?

Deepak Dalvie 22

Why is Raloxifene is Devoid of any IADRs or DDI

è  Primary route of raloxifene metabolism – Glucuronidation

SHOOH

O O

N

SHOOGlu

O O

4'Glucuronide (M1)

6-Glucuronide (M2)SGluO

OH

O O

50% reduction in covalent binding observed in the presence of GSH and UDPGA!

0

5

10

15

20

25

30

35

40

45

50

NoCofactors

NADPH NADPH +GSH

NADPH +UDPGA

NADPH +GSH

+UDPGAIncubation of Radiolabeled Raloxifene with Human Liver

Microsomes

Pmol

s of

Rad

iola

bele

d R

alox

ifene

Cov

alen

tly B

ound

to

Hum

an L

iver

Mic

roso

mes

68

72

50

% bound relative to only NADPH

Covalent Binding Incubation of [14C]Raloxifene with HLM

Glucuronidation of raloxifene

Dalvie D, Kang P, Zientek M, Xiang C, Zhou S, and Obach RS (2008) Chem. Res. Toxicol. 21, 2260.

Deepak Dalvie 23

Impact of Intestinal Metabolism on Bioactivation of Raloxifene

è  Glucuronidation primarily catalyzed by 1A8 and 1A10 l  UGT1A1 also catalyzes the glucuronidation but to a minor extent l  UGT 1A8 and 1A10 are exclusively present in the small intestine

Preincubation with Human Intestinal Microsomes results in significant decrease

in covalent binding!

0

5

10

15

20

25

30

35

NADPH only NADPH + UDPGAPreincubation of Radiolabelled Raloxifene with Human Intestinal

MicrosomesPm

ols

of R

adio

labe

lled

Ral

oxife

ne B

ound

to H

uman

Li

ver M

icro

som

es

No Cofactors NADPH NADPH + GSH

Control (no UDPGA)

Add 100 uL of this aliquot to rat or human liver microsomalincubation mixtures at

EACH TUBE CONTAINS

HIMProtein – 0.3 mg/mL - HIMNADPH – 1 mMUDPGA – 3 mMGSH – 3 mMMgCL2 – 10 mMAlamethicin – 10 µg/mg protein

WARM FOR 5 MIN

ADD SUBSTRATE 10 µM

0 min 5 min 15 min 30 min

HLM

Protein – 1 mg/mL - HLMNADPH – 1 mMUDPGA – 3 mMGSH – 3 mMMgCL2 – 10 mMAlamethicin – 10 µg/mg proteinpreincubate

INCUBATE FOR 45 MIN

1.Quench with ACN containing /internal Std

2.Determine the G1 formed

HLM HLM HLM

Add 100 uL of this aliquot to rat or human liver microsomalincubation mixtures at

EACH TUBE CONTAINS

HIMProtein – 0.3 mg/mL - HIMNADPH – 1 mMUDPGA – 3 mMGSH – 3 mMMgCL2 – 10 mMAlamethicin – 10 µg/mg protein

WARM FOR 5 MIN

ADD SUBSTRATE 10 µM

0 min 5 min 15 min 30 min

HLMHLM

Protein – 1 mg/mL - HLMNADPH – 1 mMUDPGA – 3 mMGSH – 3 mMMgCL2 – 10 mMAlamethicin – 10 µg/mg proteinpreincubate

INCUBATE FOR 45 MIN

1.Quench with ACN containing /internal Std

2.Determine the G1 formed

HLMHLM HLMHLM HLMHLM

Preincubation Experiment

Daniel C. Kemp, Peter W. Fan, and Jeffrey C. Stevens Drug Metab. Dispos. 30, 694.2002 Dalvie D, Kang P, Zientek M, Xiang C, Zhou S, and Obach RS (2008) Chem. Res. Toxicol. 21, 2260.

Deepak Dalvie 24

Efficiency of Raloxifene Glucuronidation versus Oxidation

Glucuronidation Oxidation

Cl int glu Clint oxi

µL/min/mg protein µL/min/mg protein

HIM 397 ± 14 26 + 3.1

HLM 37 ± 1.6 142 + 5.8

Efficiency of detoxification pathway explains the impact of glucuronidation on hepatic bioactivation

Dalvie D, Kang P, Zientek M, Xiang C, Zhou S, and Obach RS (2008) Chem. Res. Toxicol. 21, 2260.

Deepak Dalvie 25

What does this data tell us!

è  Intestinal glucuronidation limits the amount of raloxifene undergoing bioactivation

è  Dose of raloxifene – 60 mg QD è  Other pathways of clearance are responsible for reducing the body

burden

Liver

GI Tract

Systemic Circulation

Site of metabolism

Low exposure of Raloxifene

DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4DrugDrug

RMRM

M1M1

M2M2

M3M3

M4M4

Reduces the body burden

Highly Efficient process

Deepak Dalvie 26

Lesson!

è  The compound would have been “withdrawn from development” in the absence of all the relevant metabolism data to date

è  Understand the impact/contribution of other metabolic pathways

è  Identify enzymes responsible – maybe non-hepatic l  Especially the intestine

Bottomline Understand the Metabolism of the candidate

thoroughly prior to any Major Decisions

Deepak Dalvie 27

A Pfizer Case

Mitigating the Bioactivation Risk of 11-β-HSD-1 Inhibitor PF-915275

Deepak Dalvie 28

11-β-HSD-1 Inhibitors

Project Goal è  Develop an 11βHSD1 (11β-HydroxySteroid Dehydrogenase type-1) inhibitor

for the treatment of Type II diabetes

Cortisol (F) Active glucocorticoid

Cortisone (E) Inactive glucocorticoid

O

OH3C

H3C OH

OOH

O

H3C

H3C OH

OOH

HO

11βHSD-1NADPH

11βHSD-2NAD

11 11

á Hepatic Glucose

11βHSD1 inhibitor inhibits conversion of cortisone to cortisol

Deepak Dalvie 29

Lead Candidate in the 11-βHSD1 Program

Natilie Hosea Sajiv Nair

è  Has great physicochemical properties è  Good potency è  Pharmacokinetic Attributes

è  Key Issues from a metabolism perspective l  Risk of bioactivation l  No Structural Alerts – yet +ve signal in

RM assay

NHN NH2SO

O

CNPF-915275

Deepak Dalvie 30

Metabolism of PF-915275

Reactive metabolites trapped as Novel GSH Adducts

SNH

O O

N NH2

NH2

CO2H

NH

O

S

OHN CO2H

NH2HO2C

HNO

SO

HN

HO2C

HOSN

O O

N NH2

NH2

CO2H

NH

O

S

OHN CO2H

SN

NH

O

HO2C

Bis-conjugate Adduct Unique GSH Adduct

SHN

O O

NNH2

N Site of bioactivation and adduct formation

SNH

O O

N NH2

OH

SNH

O O

N NH2

HO

GSH

HLM/NADPH No metabolism at this site

Deepak Dalvie 31

Risk Assessment for PF-915275

è  Preliminary dose prediction was 30 – 50 mg l  Uncertainties in clearance (low) and Ceff

è  Oxidation leading to reactive metabolites l  a primary route

è  Question? l  Do other competing metabolic pathways or detoxication

of the reactive metabolite play a role?

Deepak Dalvie 32

Risk Analysis: Assessment of Reactive Metabolite Formation in LM/NADPH/UDPGA

SNH

O O

N NH2

NC

SNH

O O

N NH2SNH

O O

N NH2

HO OH

SNH

O O

N NH2

GluO

SNH

O O

N NH2

OGlu

SNH

O O

N NGlu

NC

H

M3

M5PF-00915275

GSH Adducts

[O]/GSH

UDPGA UDPGA

è  14C-PF-915275 synthesis accelarated to see the impact on

covalent binding l  Assess the covalent binding in vitro and in vivo

No oxidative metabolites observed in the absence

of GSH or UDPGA!

Deepak Dalvie Natilie Hosea

Minor Pathway

Deepak Dalvie 33

Covalent Binding Studies with [14C]-PF-915275 Using Human Liver S9

Ping Kang Sue Zhou

21

633

741928

335

30 1018116

10 30

200

400

600

800

Control NADPH only NADPH+GSH NADPH+GSH+ other Phase

II cofactors

pmol

/mg

Prot

ein

.

RatMonkeyHuman

S9 Assay

Drug Drug

RM RM

M1 M1

M2 M2

Drug Drug

RM RM

M1 M1

M2 M2

Drug Drug

RM RM

M1 M1

M2 M2

Drug Drug

RM RM

M1 M1

M2 M2

M3

M4

è  915275 – was bioactivated but: l  Metabolites were detoxified by GSH, sulfation and glucuronidation l  In vitro covalent binding near background when other co-factors

are used

Deepak Dalvie 34

Further De-risking Factors to Move PF-915275 into Development

è  Rat studies performedl  Very low binding to liver protein in vivo (0.21 pmol/mg;

Deepak Dalvie 35

Overall Conclusions From the 3 Cases

è  Covalent binding to microsomal proteins may not be predictive of toxicity l  Some refs by Obach et. al. further confirm this

è  Thorough assessment of metabolism is pivotal in early discovery

è  Important to consider the contribution of other metabolic pathways (hepatic or extrahepatic) in light of a positive signal

è  Some idea of total daily dose/exposure helps l  Get a range to see if the dose is lower than 20 – 50 mg

Ø  Use in vitro potency and clearance from HLM as first cut Ø  Refine the dose prediction from the efficacy model

è  Estimate the body burden of the reactive metabolite: l  DRM = D x fa x fm x fRM l  Gan and co-workers Chem. Res. Toxicol. 2009, 22, 690.

Deepak Dalvie 36

Is Covalent Binding or GSH Adduct Formation A Kill Shot??

è  Clearly, a positive signal by itself ‘is not’ a kill shot è  Not good predictors of IADR è  Needs to be considered as a ‘flag’ to trigger additional studies è  GSH conjugation is a detoxication pathway –need to put in right

context è  However, helps to elucidate mechanisms of bioactivation

è  Useful in circumventing the liability through iterative design

è  Liver microsomal assays do not always address all metabolic pathways for a compound

l  Use of hepatocytes or S-9 – supplemented with co-factors gives a better picture of all metabolic pathways

approaches to mitigating the bioactivation potential of ......alpidem (anxiolytic) hepatitis (fatal)...

Documents