Human Health Risk Assessment Approaches for Chemicals with … nexgen... · 2015-07-16 · NexGen Risk Assessments: _ Office of Research and Development . 3 &EPA . United States

Nex Gen

Human Health Risk Assessment Approaches for Chemicals with Limited Data Dr. David Dix u.s. EPA, Office of Research and Development

~.NIEHS~~ NlItionallnstitule of .;l Environmental Health Sciences

&EPA Unbd$bltas Environmentlll Protection AQency

NIH eM."" ICAL ~IL.NOMI(;1i CaN fllJo(

C.ieM1'~II:' "0'' D,jcA':;1:CD"T,.CI. a ND ~"C• .r:"'TIOH

Tox21 I ToxCastandI I I I

ExpoCast I I

reasing Weight of Evidence :

reening/Ranking

ILimited decision-making I I I I

I I Chemical SafetyI I I

for SustainabilitI I

ADVERSITY I I I I I

: I Regulatory decision-making

I

Agency

Tier I

Screening to Biological Pathway Altering Doses and Ultima

: Tier II . I

Tier III 100s of chemicals

tely Adversity

1O,OOOs of chemicals ,I 1000s of chemicals I

Inc

Sc

I

I

y

I

&EPA United States Environmental Protection

NexGen Risk Assessments:

Office of Research and Development 3 _

&EPA United States Environmental Protection Agency

CompTox and Chemical Safety for Sustainability: Supporting High Throughput Risk Assessments

• Understand chemical toxicity at a molecular level

• Understand using as few animal as possible

• Build predictive models • Initially screen and prioritization, eventually provide quantitative

points-of-departure

• Assess many chemicals - deal with the data gaps

_ Office of Research and Development 4

• • •

1000s/d ay

10,000s100,000s/d ay

.. , ..... . . .~..... ",.

1Os-1 OOs/yr

Human Relevance~

Cost/Complexity:

.... ........ .... . ......... ,, ~.. • <&

Gene-expression

.. ...

Throughput/

High-Throughput Screening Assays batch testing of chemicals for pharmacologicalltoxicological endpoints using automated liquid handling, detectors, and data acquisition

&EPA United States Environmental ProtAgency

ection

Simplicity Office of Research and Development _


High Throughput Screening 101

Chemical Exposure

Assay Target Biology (e.g., Estrogen Receptor)

96-,384-, 1536 Well Plates HTS Robotic Platfonn

Office of Research and Development _


Tox211ToxCast Data Analysis

~ssys~ • I I 1

. , -

_ : Assay-Chemical Hit

..,•

• •

•

Office of Research and Development 7 _

&EPA ToxCast: Multiple Targets per Pathway United States Environmental Protection Agency

~]J in P.CGF

rr"WI smr1~ • ! 'tEKlracellular rr:-::rm I'I\H!,

Inwli , rnAF~Cytoplasm ~I ~ S.3>""r~S2S"'C:r':119 ~

• M ,'\ I,• Biologically Multiplexed PPARi5 Li~and~""T?fdActivity Profiling (BioMAP) ('"4"

{t.-Ht + 1

o Multiplex Transcription WCK112 11,.0( ,,~ ~Reporter Assay

'"' cx~, _ =lSP':l~,~- --tt.-~

r@' ~_ ~ "P R6 Li~:oo

-~o Cell-based HTS Assays ~ Nucleus• Cell-free HIS Assays

a High Content Cell Imaging Assays

- ~STATS-STATS.- ~-

,~~l.ooIPy "~>l.J'- .rollr~ "IiUII 11.Jpld -<>moo,la:i.

~ ,roy flr <lm;c,. en me ~'~""'"~,,n1 ~.SI-.",pollr~ .Ii~":!- U,i<J"", "",liljs 4 '_~"In· r.f"lfn"!J~""oJ~

4 Hep91oc,..rir>;JG""'" ~ I An rx>r'fP. ,ffArM'''' m ::I.G U:O,. ~ '" I',U If' h""",,,l"'''

:' M",-"f'h'iJ" tllnr!lcY

Office of Research and Development

8

_

CRTC2 ~~~---.,:fOX01 .

n ~ 1TCF7L2 ~RE~ ~P:KA=;, ... GCK ~ :1f prolNS ~ Insulin

PDX1 et al :! PC1/3 ~ b-cell V\IV'~/~/~;:';\ ... differentiation

it' ~ Foxa2 Fgf /

... Fgfr2 _ Ngn3 _Hnf1b A. NIEHS

------....;;a~ ~~ H.tIonlllnstitut.of&'tlronnwRtloIH8IIJdtSct

I ...

b-cell mass expansion....( # V· .nc

Glut2

1~~~Aignalingin Pancreatic ~-Cells

United States ""'lEi---------Environmental Protection • InsulinA,,,,y IGF~1 ,-lnStUlinIGF1R llNSR

K'+Ca"Wnt Glc

0

~ Frizzled PI(3,4,5)P3

i GIUt2! lK' channel .1

Ca" ..'~ PD~I !~

GLK

Glc6PDishevelled K'.,PTEN t Insulin1 @~ Ca"., adenylyl_~~...,.;;;.~ A!SI [ATP] cyclase

'-Gsk3b to

F [cAMP]

]l-cateni

fficeO

M

_


Sample Output From Signaling Hyperlinks to ToxCastDB

GSK3b You ue h.,e: EPA Hom•• Nation.1 C,nW for Compu''''on.1 Toxicolo~y' ToxC"tre- "'".y

c;c;;J ToXlofDI11 ToxCu.1>I II Ex~oC.<tDlI D'H.xDI I H,me 18M;' Info I D,t, Colle;~on U,t I Chem~,1 Li,j I (lenM A"o,i.ted ..th A",y, I Help

Assay: Novascreen Human GSK3b AS'iiIYl!I: SO'.~'

SOl. ce Na",e All

,~. No,.""," Hun.n GIK3b

Deicriplioll Hun.n GSK3b fluo""".in-p.ptid,

N",,,IJ., of S'~l5'"n""s m

Hu"lle, or CO'fIIIO""'~S

Sp.cies Hano ",i,n,

Parameters

CATALOG NUMBER 20)_0125

ASSAYCATEJORY Et1lj'tl1e I1hibiton

~SSAYCATE30RY In ~tro (Eiochellical) ASSAYTARGET GeK3b

~88!1YTMGET f!l~ILY Kinooc

ASSAYTARGET SCURCE Rocomb nant

ASSAY GENE ID 2932

ASSAY GENE NAME GeK38

,o.oOMT[CI tWLCOY rlcor.".n,. -D~O

ASSAY REFERENCE CO~PO~D St,uro",orine

ASSAY NOTE Ki~""SE

",SSAYSUBSTRATE NAIIE C~GC OIOUP A88AYATP CONCENTRATlON (~) NCCT_¥1

ASSAY LIGAND NA~E 1, E-06

ASSAY LIGAND C~CENTRATlON (~) 1.'OE-O' ~SSAY8wv( Filorescein-p,ptide' ATP -> Fluorescein-phosphopeptiJe' KlP

[-::-_-::-=::,0:::':::"::-=-=-::-:::Nil"'. CASRN NVS_[NZ_hCSK3b (lIM\ nurn ber of "actives" =3 Nanco,eb 80H-01-7 0.27

Naneb 124,7-38-2 0.32

CREB Assay: Attagene Factorial cis eRE Ass~ld; "Source All:>pv>e

Source ll.lllle AID ATC_CRE-CIS

ll.lme All:>9*N F~clorl~1 ell CR[

DesCllption f~c,,,,",1 ,.port., go... uoay

llumbel of SubsUnces m ,llumbel of Comllonents Slledes Homo oap~nt:

Parameters Puamete. Vallie

ill Out nIT _._ASSAYUil.

ASSAY CATEGORY In vlro (Celuler)

ASSAY TAROET cAMP Re$pOl"lSe R1etn':!rf

ASSAY TAROET fAMLY Tr8l1sc,jp:ioo Factor

ASSAY TAROET SOURcE Cellne

ASSAY TAROET SOURCE TYPE ASSAY 0Ef\E () """" ""'" ASSAY 0Ef\E NAAE CREBJ ASSAY TEOf\lOLOOY Reporter gene a::::ay

ASSAY~ DNA ::equerlCer

ASSAY REfERENCE COMI'OLN:l Forskoln cANP

ASSAY f>KITE "hW:~xed 'e~er gene aSSay, cAMP, cGMP, f>KI receptor, GPCR plllhwlltYs·

lIame-""''''' Az~·melh~

Azoxyslrobit

BendiocIllfD

Bisphenoj A

a_"",

"""~"'"'" CyazofMlid

OW.... D~zomet

AIellYn (d-ei::,tr8l1S)

f"lr.Hr..-1VI

CASRlI

15972-60-8

101-05-3

86-5o.<J 1311360-33-6

22781-23-3

0).05-' 1689-84·5

101-Zl-3

120116-88-3

1Zl552-61_2

533-74-4

584-79-2

!N-::IO-!l

ATG_CRf_CIS (tiM) ,. 59.0

27.0

".0 51.0 partial list: 300 number of total "actives" = 52 40.0

~.O

10.0

730

49.0 46.0

4.1 n Nellram-Ilnc 900(-42-2 I~.O

http://ador.epa.qovlactor/facesfToxM inerIH om e. isp

•

adenylyl cyclase

GSK3b

Fox01

FoxA2

PTB1b

&EPA United States Environmental Protection Agency PTEN AKT

TM ToxPi for Insulin Signaling in IGF1R

Pancreatic J3 Cells- Top 30 from W channels l.......~

309 Chemicals in ToxCast Phase I

CREB

Metirarn-zinc Mancozeb Maneb PFOS Cyclanilide 3-1000-2- Oxytetracycline Haled Berlomyl propynylbutylcarbamate dihydrate Tebufenpyrad----------------------'

Methylene bis(thiocyanate) Chlorothaklflil Niclosamide TCMTB Imazalil Spiro)lsmine Milbemectin Cyprdinil DiqU8tdibromide Fludioxonil

sethoxydim Trichlorfon Forchlorfenuroo d-cil, tnlnl BisphenolA PJopyzamide Anilazine Chlorpropham Captan Dichlaran Allethrin

Office of Research and Development http://epa.gov/ncctfToxPi/

'

_


ToxCast and Tox21 Assays and Chemicals

500--.---

Phase I

Assays

50 -+- _______ Tox21 ---j

300 1000 10,000 Chemicals



What isHigh-Throughput Risk Assessment?

• Where does risk assessment come in?

- Estimate upper dose that is still protective

-RfD, BMD, POD

• Where does high-throughput come in? - Focus on molecular patnways and targets whose perturbation

can lead to adversity

-Screen hundreds to thousands of chemicals in in vitro assays for those targets

-Get oral dose using H-T pharmacokinetic modeling

• Incorporate population variability and uncertainty



Why do we need High Throughput Risk Assessment (HTRA)?

• Thousands of chemicals with no or little animal data

• Need starting points for setting health-protective exposure levels

• These starting points can be used to prioritize and target further testing


&EPA United States Environmental Protection Agency HTRA Basic Outline

1. Define molecular pathways linked to adverse outcomes

2. Measure activity in vitro in concentration-response (PO)

3. Estimate external dose to internal concentration scaling (PK)

4. Estimate dose at which pathway is perturbed in vivo

5. Estimate population variability and uncertainty in PK and PO

6. Estimate lower end of dose range for perturbation of pathway



HTRA Publication

Chemical Research in Toxicology, in press, 2011

Estimating Toxicity-Related Biological Pathway Altering Doses for High-Throughput Chemical Risk Assessment

Richard S. Judson, Robert J. Kavlock, R. Woodrow Setzer, Elaine A. Cohen Hubal, Matthew T. Martin, Thomas B. Knudsen, Keith A. Houck, Russell S. Thomas, Barbara A. Wetmore, David J. Dix

National Center for Computational Toxicology, Office of Research and Development, U.S. Environmental Protection Agency

The Hamner Institutes for Health Sciences, Research Triangle Park, North Carolina



What Pathways to Use?

• Start with known targets (genes, proteins) and pathways

• Define levels of evidence and prioritize for analysis - Class 1 - the link between in vitro activity and adversity is clear

(e.g. cholinesterase activity). There is asingle target which, if significantly perturbed, can lead directly to undesirable phenotypic change.

- Class 2 - here is an association between perturbations of a pathway and some disease outcome, but the details and causal linkage is not clear (e.g. PPAR pathway perturbations and potential linkage with human disease).

-Class 3 - no clear linkage between in vitro activity and adverse in vivo outcomes is currently known.


&EPA United States Environmental ProtectioAgency

Measuring the Pathways n

• ToxCast and Tox21 are using hundreds of assays on thousands of chemicals

• Need to determine concentration at which pathway is "altered"

• Many ways to do this -Simple - take minimum AC50 (AC20, etc.) of any assay mapping

to the pathway

- Harder - develop a systems-level model of the pathway and build a probabilistic concentration-response profile

• Add in estimates of population variability and uncertainty


• •

• • •

- - - -

N '"

0 N

~ 0 c -'" 0 Co 0

0: ~

0-

- ~spheool A ACSO: 1 1 •

•

ATG ERa TRANS

o. , 0.5 50 50.0

Concentration fuMl

- SIspMnol A AC50: 0.64

•

ATG ERE CIS

0.1 0.5 5.0 50.0

Concentration luM\

&EPA Example: Concentration-Response United States Environmental Protection Agency Curves for Bisphenol A

Sample curves for SPA in two of six ToxCast ER assays

Note that full concentration-response profiles can be measured, at arbitrary spacing and to arbitrarily low concentrations (at moderate cost for a given chemical)


~ nited States nvironmental Protection gency

Significance o ~of Reverse t:

~ 0

~ ..... Q) u c oicokinetics:

- Cl)

u <t:ing Kinetics ()

fil

Critical to ?N

f') o 0° 8 o

, , , •,

.J.

• I

•I

I

• I I

• I

~

I. ... I I

.J.

~derstanding ~

ynamics N o +CD .....

o o +(I) .,

N o I(I)

o ~

D Estimated Exposure I o 8 ;gO

on 0 8 ~ iaDt~ o ~ 8 ~ ~ ~ ~ ~ r B~8 ::

o 0" Q T , I ,.L :: :§~ G'nerig 6B O a ~ ~ B~~T 8'"~gD"-0 ~~tjI II I ~ ~ II : I :51 I I II -'-:I ::I I

O~QgOT I. I: I 10 1 I~ t·~ I I ~ ¥ ..L J.. .J. .L I I I I I I I ...l.."'0'" I ..L I I I I I I IT L..ri I I I I I I I t I •: I8 I : I I I C ~ I , ....I... ..... I:

I I ,~~ C I' ~ ~ I ..L I I I I: I I 0 l' I .L -'- .J.I I I I .Le nilI .L I

I I...l.. I IJ [J

ClB~ .L DclJ

lJ D

D Il I

D D-'

Office of Research and Development <b -'-------,.------,11-.-'-.-'--.1----,,---,',--.-1-.- , .-,-'- I' -'-,-'-,-'-,----',---'I:-r- ,.-----...-- 20I,.---D-,.--,-"-1-'-,--',--',----'I,--r- ,"------'-.-,-.-,--.,--"--'I~r'-r-,.--,,--,.---,, .-,-'- ,,----'I I

Ratroff, et al. Tax.Sci. 2010

I

q,E· UEA

Theo

ToxAdd

isUn

D

_

I Adverse Effect

MOA'\. 1

Key Events

1 Toxicity Pathway

BPADL

----• --• •Probability Distribution !!!!!! for Dose

HTSAssays that Activates

! Biological Pathway

Biological Pathway Activating Concentration (BPAC) Probability Distribution

Dose-to-Concentration Scaling Function (Cgg/DR)

Probability Distribution

" Populations PK Model

Plasma Protein Intrinsic Binding

&EPA Reverse Toxicokinetics (rTK): United States in vitro concentration to in vivo dosee ..... · ~---~ .. ~I .... ~-~ ~.~~ ••

Pharmacodynamics Pharmacokinetics

Clearance -


Uncertainty and Variability

• RTK modeling explicitly incorporates human population variability in PK (SimCyp)

• Other uncertainty and variability ...

- PK uncertainty due to model and data uncertainty - PO variability due to intrinsic variability in enzymes, receptors,

pathways

- PO uncertainty due to details of assay performance, etc.

• Need to develop approach to move away from using defaults for HTRA

- Follow similar path to what is being developed for standard RA



Example: Bisphenol A Estrogenicity In Vitro vs. In Vivo Reproductive Toxicity

• Rat reproduction tests resulted in a No Effect Level of 50 mg/kg/day

• Adjusted for uncertainty and variability, the no effect dose is 0.5 mg/kg/day

• HTRA lower limit BPADL99 is 0.16 mg/kg/day, derived from six ToxCast estrogen receptor assays


Estrogenicity

Lowest AC50 JUJUfrom 6 ToxCast HTSAssaysER assays

BPAC= 0.64 IJM A

Biological Pathway Activating Concentration (BPAC) Probability Distribution

Pharmacodynamics

( Adverse Effect ".

MOA"", 1 \ Key Events

TOXicity!athWay _.. -~

• ;&-, ..., i

Pharmacokinetics

" Populations PK Model

Plasma Protein

Dose-to-Concentration .-r- Scaling Function (Css/DR)

Pro7:\:ution

= kg/day

NEL/100= 0.5 mg/kg/day

BP~

Probability Distribution

BPADgg

0.44 m

for Dose that Activates

Biological Pathway

I~::::;"---__ Estrogenicity

BPADLgg = 0.16 mg/kg/day

Intrinsic Clearance

L....

Binding Exposure0.00008

=

mg/kg/day

&~~~s . Example: Bisphenol A Estrogenicity In Environmental Protection Agency Vitro vs. In Vivo Reproductive Toxicity


HTRA Summary

1, Select Toxicity-related pathways 2. Develop assays to probe them 3. Estimate concentration at which pathway is "altered" (PO) 4. Estimate concentration-to-dose scaling (PK)

5. Estimate PK and PO uncertainty and variability 6. Combine to get BPAD distribution and safe tail

• Many (better) variants can be developed for each step (1-6)

• Use for analysis and prioritization of data poor chemicals



HTRA SUmmary (2)

• Pathway perturbation = MOA Key Event evidence

Necessary for MOA

- Sets lowest dose at which chemical acts through MOA

- Do not need to do low-dose extrapolation - just measure it


&EPA NexGen Risk Assessment Data Requirements United States

~~~;~~m.""IP'O"";O" Being Met by CompTox and CSS Research

Tier I 1O,OOOs of chemicals

, Ultra-High Throughput I

Screening (uHTS) for II

Bioactivity Profiling- I

Tox21

• In vitro assays (dozens) • In silico assays

./QSAR

./Molecular Docking • ToxPi (Toxicological

Prioritization index) data integration

./Pathway, toxicity or disease associations

Increasing Weight of Evidence Screening/Ranking

Office of Research and Development

I I I

Tier II 1000s of chemicals

HTS, Metabolism, Dosimetry, Exposure, Modes ofAction and Systems Integration-

ToxCast and ExpoCast

• In vitro & in silico assays (hundreds)

• Metabolism and dosimetry • Exposure estimates • Alternative species, systems

modeling, virtual tissue simulations

• High Throughput Risk Assessments (HTRA) based on modes of action

I I I I I I I

Tier III 100s of chemicals

High Content, Lower Throughput, Animal or

Human Studies-Chemical Safety for

Sustainabilitv

• Molecular epidemiology & clinical studies

• Exposure modeling and biomarkers of exposure

• Molecular biology & targeted animal testing

• Pathway and disease outcomes

• Knowledge-Base integration and visualization

Adversity I I

Limited decision-making I

I Regulatory decision-making -


Acknowledgements

• Participants in the Day 3/Tier 2 portion of Nov201 0 NexGen Workshop: Stan Barone (EPA), Derek Knight (ECHA), Karen Leach (Pfizer), Chris Portier (ATSDR), Mike Devito (NTP), Alex Tropsha UNC), Richard Judson (EPA), David Reif (EPA), Rusty Thomas (Hamner), Jason Lambert and Iia Cote (EPA)

• Tox21 teams at EPA, NIEHS/NTP, NHGRI/NCGC and FDA: led by Bob Kavlock, Ray Tice, Chris Austin and David Jacobson-Kram

• ToxCasUExpoCast EPA teams: led by Keith Houck and Elaine Cohen Hubal

• Newly forming Chemical Safety for Sustainability teams at EPA


Documents

Human Health Risk Assessment Approaches for Chemicals with … nexgen... · 2015-07-16 · NexGen Risk Assessments: _ Office of Research and Development . 3 &EPA . United States