2015 NAS Interindividual Variability Workshop | September 30, 2015 Integrating In Vitro and In...

2015 NAS Interindividual Variability Workshop | September 30, 2015

Integrating In Vitro and In Silico Methods to Evaluate Variability

Barbara A. WetmoreThe Hamner Institutes for Health SciencesResearch Triangle Park, NC USA [email protected]

mailto:[email protected]


Broad-Based Movement in Toxicology Towards In Vitro Testing and Hazard Prediction


HT Toxicity Testing DataDifficulty Translating Nominal Testing Concentrations

into In Vivo Doses

Knudsen et al. Toxicology 282:1-15, 2011


Prioritization and Hazard Prediction Based on Nominal Concentrations Can Misrepresent Potential Health Risks

Protein Binding

Metabolic Clearance

Bioavailability

van de Waterbeemd and Gifford, Nat Rev Drug Disc 2:192, 2003

Reif et al. Environ Hlth Perspect 118:1714, 2010


Incorporating Dosimetry with High-Throughput Screening Data

Human Hepatocytes

(10 donor pool)

HumanPlasma

(6 donor pool)

-5

-4

-3

-2

-1

0

1

2

3

0 50 100 150

Ln C

onc

(uM

)Time (min)

Nifedipine

1 uM initial

10 uM initial

Hepatic Clearance

Plasma Protein Binding

Population-Based In Vitro-to-In Vivo Extrapolation

Plasma Concentration at Steady State for 10,000

Healthy Individuals of Both Sexes from 20 to 50 Yrs Old

Setting the Stage…


Using Reverse Dosimetry to Estimate Population-Based Oral Equivalent Doses

Upper 95th Percentile Css Among 10,000 Healthy

Individuals of Both Sexes from 20 to 50 Yrs Old

Plasma Concentration

ToxCast AC50 Value

Assay X (e.g., ACE inhibition)

Oral Exposure

Oral Dose Required to Achieve Steady State

Plasma Concentrations Equivalent to AC50

Oral Equivalent (mg/kg/day)

ToxCast AC50 (uM)

1 mg/kg/day

Upper 95th Percentile Css (uM)

=

1 mg/kg/dayOral Exposure

Metabolic and binding parameters

Setting the Stage…


Incorporating Dosimetry and Exposure with HTS Data


Primary Hepatocytes

-5

-4

-3

-2

-1

0

1

2

3

0 50 100 150

Ln C

onc

(uM

)

Time (min)

Nifedipine

1 uM initial

10 uM initial

Hepatic ClearanceCl in vitro Plasma Css

General Population

≠

Plasma Css

General Population

Dosimetry and Exposure Strategy Limited to General Population


The Impact of Population Variability on Risk Assessment

Clearance results for full database (27 substrates).

from Ginsberg et al., 2002, Toxicol. Sci., 66, 185-200.

Rat

io to

you

ng a

dults

Age (years)

from Ginsberg et al., 2005, Environ. Health Persp., 113, 1243-49.

Clearance differences span acrossmultiple newborn and juvenile life stages…

… as well as geriatric life stages.


Percent contribution of individual CYPs to total hepatic clearance of xanthines.

Ginsberg et al., 2004, J. Toxicol. Env. Health Pt. A, 67:297-329.

The Impact of Population Variability on Risk Assessment

Cresteil, 1998, Food Addit. Contam., 15:45-51.

Ontogeny of Human Hepatic CYP Isozymes Frequency for CYP2D6 alleles classified as functional, non-functional and reduced functioning for various subpopulations.

Bradford, 2002, Pharmacogenomics. 3(2):229-43.

Sole reliance on pharmacokinetic data for a “generic” population could lead to a significant underestimation of risk

to a susceptible lifestage or subpopulation


CYP1A2

CYP2D6

CYP2C8

CYP2E1

UGT1A4

CYP3A5

CYP3A4

CYP2C19

UGT1A1

CYP2C9

CYP2B6

UGT2B7

Population-based In Vitro-In Vivo Extrapolation

Primary Hepatocytes

-5

-4

-3

-2

-1

0

1

2

3

0 50 100 150

Ln C

onc

(uM

)

Time (min)

Nifedipine

1 uM initial

10 uM initial

Hepatic ClearanceCl in vitro Plasma Css

General Population

CYP1A2

CYP…

CYP3A4

UGT…

ClrCYP1A2

ClrCYP…

ClrCYP3A4

ClrUGT…

Plasma Css for:

Neonates

Northern Europeans

Asians

Children

And so on…

Intrinsic Clearance Rates

rCYP1A2

rCYP3A4

rCYP…

rUGT …


Integrating High-Throughput Pharmacokinetics with the ToxCast In Vitro Assays

Reverse Dosimetry

Oral Exposure

Plasma Concentration

ToxCast AC50 Value

Oral Dose Required for Specific Subpopulations to

Achieve Steady State Plasma Concentrations

Equivalent to In Vitro Bioactivity

~600 In Vitro ToxCast Assays

Ora

l Equ

ival

ent

Dos

e (m

g/kg

/day

)

What are humans exposed to?

?

?

?

Chemical

In VitroBioactivity

Recombinant Enzyme

MetabolismHuman Plasma Protein Binding

Population-Based IVIVE Model

Steady State Plasma Concentrations for Different

Lifestages or Subpopulations

Least Sensitive Assay

MostSensitive Assay

Wetmore et al., 2014, Toxicol Sci, 142(1):210-214


0 20 40 60 80-4

-3

-2

-1

0

1 CYP3A4 CYP3A5 CYP2C19CYP2C8CYP2C9CYP1A2CYP2C18

Time (min)

Ln

Co

nce

ntr

atio

n (

uM

)

0 20 40 60 80-0.4

-0.2

0.0

0.2

0.4

0.6

UGT1A4

Time (min)

Ln

Co

nce

ntr

atio

n ( M

)

0 20 40 60 80-10

-8

-6

-4

-2

0

2CYP2E1CYP2C9CYP1A2CYP2C19CYP2B6

Time (min)

Ln

Co

nce

ntr

atio

n ( M

)

No UGT metabolism detected

Carbaryl Difenoconazole

Recombinant Isozyme Clearance Rates


Incorporating Recombinant Phase I and II Enzyme Data into IVIVE Modeling

Clint = intrinsic clearanceISEF = Clint, HLM

Clint, rCYP x HLM CYP abundanceHLM = human liver microsomes

rCYP = recombinant CYP isoform(pmol P450 / mg protn)

(dimensionless)

(uL / min / mg protn)

(uL / min*pmol P450)

12.25

3.8

5.4

16.5

0.2

3.41.9

140.3

33

4.3 1A22A62B62C82C92C182C192D62E12J23A4

Hepatic CYP Isozyme Abundance in Healthy Adults

(% of Total)

Scaling rCYP Data to HLM using intersystem extrapolation factors


Combining Isozyme Clearance and Abundance Data to Determine Fraction Metabolized

Isozyme No. Chemicals% fm > 5%

% fm Range Chemicals with% fm > 5%

CYP1A2 3 0.4 - 91.4 Bensulide, Carbaryl, Fludioxonil

CYP2C9 6 2.1-63.1 Azoxystrobin, Bensulide, Carbaryl, Difenoconazole, Haloperidol,

Tebupirimfos

CYP3A4 7 1.0-80.2 Acetochlor, Azoxystrobin, Bensulide, Difenoconazole, Haloperidol, Lovastatin

Tebupirimfos

CYP3A5 2 1.4-6.4 Lovastatin, Tebupirimfos

UGT1A1 2 2.6-19.3 Haloperidol, Tebupirimfos

UGT1A4 3 0.1-12.1 Difenoconazole, Haloperidol, Lovastatin


Carbaryl

0.01

0.1

1

Cs

s a

t1

M (

M)

(5th

-95t

h %

ile)

Fludioxonil

0.1

1

10

Cs

s a

t1

M (

M)

(5th

-95t

h %

ile)

Comparison of Css Values Derived Across Multiple Lifestages and Subpopulations

Upper 95th percentile Css

Lifestage or Subpopulation(Age (yr) or Ethnic)


Health

y

Chines

e

Japa

nese

North

Europ

ean

Diabet

ic

Renal

Insu

fficien

cy6-

20 2-6

0.5-

20-

0.5

Health

y

Chines

e

Japa

nese

North

Europ

ean

Diabet

ic

Renal

Insu

fficien

cy6-

20 2-6

0.5-

20-

0.5

HKAF =11.4 HKAF =11.5


Difenoconazole

0.01

0.1

1

Cs

s a

t1

M (

M)

(5th

-95t

h %

ile)

Acetochlor

0.001

0.01

0.1

1

Cs

s a

t1

M (

M)

(5th

-95t

h %

ile)




Health

y

Chines

e

Japa

nese

North

Europ

ean

Diabet

ic

Renal

Insu

fficien

cy6-

20 2-6

0.5-

20-

0.5

Health

y

Chines

e

Japa

nese

North

Europ

ean

Diabet

ic

Renal

Insu

fficien

cy6-

20 2-6

0.5-

20-

0.5

HKAF =3.5

HKAF =6.7


Wetmore et al., 2014, Toxicol Sci. 142(1):210-214.



Agreement between In Vivo and IVIVE-derived Css Values using Recombinant

CYP-based Clearance Rates

ChemicalIn vivo PKCss (mM)

IVIVECss (mM)

Carbaryl 0.030 0.046

Haloperidol 0.090-0.126 0.029

Lovastatin 0.004-0.009 0.001


Estimated Chemical-Specific Toxicokinetic Adjustment Factors

Chemical Median Css for Healthy Population

95th Percentile

Css for Most Sensitive

Most Sensitive

Estimated HKAF

% Contribution of Isozyme

Differences to Average HKAF

Acetochlor 0.026 0.15 Neonatal 6.7 86

Azoxystrobin 0.099 0.66 Neonatal 6.7 86

Bensulide 0.241 0.97 Neonatal 4.0 79

Carbaryl 0.043 0.49 Neonatal 11.4 87

Difenoconazole 0.201 0.49Renal

Insufficiency 3.5 99

Fludioxonil 0.38 4.37 Neonatal 11.5 87

Haloperidol 0.029 0.14 Neonatal 4.9 83

Lovastatin 0.001 0.009 Neonatal 6.5 90

Tebupirimfos 0.107 0.38Renal

Insufficiency 3.5 15


Acetochlor

<1 1-2

3-5

6-12

Femal

es

20-5

0

50-8

00.0001

0.001

0.01

0.1

1

10

100

1000

10000

100000

Ora

l Eq

uiv

alen

t D

ose

or

Est

imat

ed E

xpo

sure

(m

g/k

g/d

ay) Difenoconazole

<1 1-2

3-5

6-12

Femal

es

20-5

0

50-8

00.0001

0.001

0.01

0.1

1

10

100

1000

10000

100000

Ora

l Eq

uiv

alen

t D

ose

or

Est

imat

ed E

xpo

sure

(m

g/k

g/d

ay)

♦●▼ ■▲■▲ ♦ ӿ

Matching Oral Equivalent Doses and Exposure Estimates for Different Populations

●ӿ

▼♠ ♠

Subpopulation or Lifestage(Age (yr) or Ethnic)

Subpopulation or Lifestage(Age (yr) or Ethnic)


Matching Oral Equivalent Doses and Exposure Estimates for Different Populations


Incorporating Population TD Variability with PK Variability

Abdo et al. 2015, Environ Int Sep 17; 85:147-155.


Key Points

• First comprehensive attempt to combine physiologic and PK differences to quantitate variability anticipated between life stage-, ethnic- and disease-based populations.

• First effort to evaluate PK variability in a manner that could 1) identify sensitive populations and 2) replace use of default safety factors in risk assessment.

• Subpopulation- or lifestage- based pharmacodynamic differences will also contribute to the variable susceptibilities that may be observed following chemical exposure.

• Tools are emerging that allow the incorporation of pharmacokinetic with pharmacodynamic variability.


Acknowledgements

Hamner - Institute for Chemical Safety Sciences

Rusty Thomas (EPA-NCCT)John Wambaugh (EPA-NCCT)Lisa M. Almond (Simcyp)Masoud Jamei (Simcyp)

External Collaborators

American Chemistry Council –Long Range Initiative

Funding

Brittany AllenMel AndersenHarvey Clewell Alina EfremenkoEric HealyTimothy ParkerReetu SinghMark SochaskiLonglong Yang

2015 NAS Interindividual Variability Workshop | September 30, 2015 Integrating In Vitro and In...

Documents

Transcript of 2015 NAS Interindividual Variability Workshop | September 30, 2015 Integrating In Vitro and In...