From Data to Action: Bridging Chemistry and Biology with Informatics at NCATS
-
Upload
rguha -
Category
Data & Analytics
-
view
134 -
download
0
Transcript of From Data to Action: Bridging Chemistry and Biology with Informatics at NCATS
From Data to Ac+on Bridging Chemistry and Biology
with Informa+cs at NCATS
Rajarshi Guha, Ph.D.
BCHB-‐597 February 2015
What is Transla+on?
Transla'on is the process of turning observaEons in
the laboratory and clinic into intervenEons that
improve the health of individuals and the public -‐
from diagnosEcs and therapeuEcs to medical
procedures and behavioral changes.
What is Transla+onal Science?
Transla'onal Science is the field of invesEgaEon
focused on understanding the scienEfic and
operaEonal principles underlying each step of the
translaEonal process.
NCATS studies transla1on as a scien1fic and
opera1onal problem.
NCATS Mission
To catalyze the generaEon of innovaEve methods and technologies that will enhance the development, tesEng and implementaEon of intervenEons that tangibly improve human health across a wide range of human diseases and condiEons.
NCATS Scien+fic Ini+a+ves • Clinical Transla1onal Science
– Clinical and TranslaEonal Science Awards – Rare Disease Clinical Research Network – New TherapeuEc Uses program
• Preclinical Transla1onal Science – NCATS Chemical Genomics Center – TherapeuEcs for Rare and Neglected Diseases program – Bridging IntervenEonal Development Gaps program
• Re-‐engineering Transla1onal Sciences – Toxicology in the 21st Century – Microphysiological Systems (Tissue Chip) program – Office of Rare Diseases Research
Preclinical Development/TRND
BrIDGs
FDA CollaboraEon
Systems Toxicology (Tox21)
RNAi
Paradigm/Technology Development
Repurposing
Lead Optimization
Preclinical Development
Probe/Lead Development
Target Validation Target
FDA approval Clinical Trials
I II III
Project Entry Point
Deliverables
Repurposing
Unvalidated target
Validated target
Lead compound
Preclinical development candidate
Genome-‐wide RNAi systems biology data
Chemical genomics
systems biology data
Small molecule and siRNA research probes
More efficient/faster/cheaper translaEon and therapeuEc development
Leads for therapeuEc development
PredicEve in vitro toxicology profiles
Approved drugs effecEve for new
indicaEons
New drugs for untreatable diseases
Novel clinical trial designs
Drugs suitable for adopEon for further
development
Assay Dev
Assay , Chemistry Technologies
Target assay
DPI Program
NCGC/Probe Dev
NCATS DPI: A Collaborative Pipeline NCATS DPI: A Collabora+ve Pipeline
NCATS Chemical Genomics Center • Obligatory collaboraEon model
• Currently > 250 collaboraEons with invesEgators worldwide
• Assay development, HTS, chemical informaEcs, medicinal chemistry: “target to POC”
• Focus is unprecedented targets, rare/neglected diseases
• Mission Ø Chemical and siRNA probes/leads Ø New technologies/paradigms to
improve efficiency and success rates of target-‐to-‐lead stage of drug development
Ø Chemical genomics: general principles of siRNA acEon, small molecule – target interacEons
NCATS Chemical Genomics Center
NCATS MISSION
• To catalyze the generaEon of innovaEve methods and technologies that will enhance the development, tesEng and implementaEon of intervenEons that tangibly improve human health across a wide range of human diseases and condiEons.
NCGC MISSION
• To develop chemical probes that fundamentally change our understanding of the molecular basis of disease, provide chemical biology tools able to validate new therapeuEc and disease management approaches and catalyze innova+ve transla+onal research.
Range of screening assays performed
Phenotype (Image-based
HCS, GFP, etc)
Pathway (Reporters, e.g.,
luciferase, β-lactamase)
Protein (Enzyme readouts, interactions, etc)
Extent of reductionism
NCGC Highlights
Paradigms & Technologies
Transla1onal Events
qHTS Paradigm
Chemical Genomics Drug Repurposing
2012 2006 2007 2008 2009 2014 2010 2011 2013 2005
Novel Probes Novel Biology
NCATS Pharmaceu1cal Collec1on Combina1on Screening
Novel Combina1ons Novel Insights
What is Transla+onal Bioinforma+cs?
• From the AMIA – “… the development of storage, analytic, and interpretive methods to
optimize the transformation of increasingly voluminous biomedical data into proactive, predictive, preventative, and participatory health. Translational bioinformatics includes research on the development of novel techniques for the integration of biological and clinical data and the evolution of clinical informatics methodology to encompass biological observations. The end product of translational bioinformatics is newly found knowledge from these integrative efforts that can be disseminated to a variety of stakeholders, including biomedical scientists, clinicians, and patients.”
• PLoS “Book” on TranslaEonal BioinformaEcs gives an idea of the range & variety of topics under TBI
What is Transla+onal Bioinforma+cs?
• Methods and data that allow you move along the scale from molecular to phenotypic to clinical – Not necessarily along the whole sequence
• Methods that enable us to go from raw data to making decisions in – Chemistry – what compound should we synthesize next? How should this compound be modified?
– Biology – is this target relevant to the disease? Is this target differenEally modulated under this condiEon?
– IntegraEve – can we gain insight by linking these data sources or data types?
RNAi Screening at NCATS
• Perform collaboraEve genome-‐wide RNAi (siRNA) screening projects (assay dev, screening, validaEon)
• Advance the science of RNAi screening and informaEcs.
• Populate a public, large-‐scale siRNA screening database. • Explore new technologies for the illuminaEon of gene funcEon.
Range of Assays!
Pathways (Reporter assays, e.g., luciferase, β-lactamase)!
!
Complex Phenotypes (High-content imaging, cell cycle, translocation, etc)!
!
Simple Phenotypes (Viability, cytotoxicity, etc.)!
!
• Cancer – Drug Enhancer/Resistance Screens
• Immunotoxins • TOP1 Poisons • Platinum Drugs/Drug Resistance • Kinase Inhibitors
– Molecular Targets in Cancer • Ewing Sarcoma • Rhabdomyosarcoma • Neuroblastoma • Breast Cancer • Melanoma • Head and Neck
– Cancer-Related Pathways • NF-κB • BRCA2-Mediated Tumorigenesis • Tissue Remodeling
• Infectious Disease – Viral Infection and Replication
• Poxvirus • Respiratory Syncytial Virus • HIV • Cytomegalovirus • Ebola Virus
– Immune Response • Other Disease-Related Phenotypes
– Parkinson’s Disease – Spinal Muscular Atrophy – Lysosomal Storage Diseases – Neuro-protection
• Fundamental Cell Biology – DNA Replication – Reprogramming/Differentiation
• And many more!
RNAi Screening at NCATS – Project Areas RNAi Screening at NCATS – Project Areas
Small Interfering RNA (siRNA)
siRNAs provide an excellent way to conduct gene-specific loss of function studies."
Correla+on – The Good News
Under optimized conditions, siRNAs yield highly reproducible assay responses."
R = 0.92
Reproducibility-‐ Same siRNA in Replicate
Z-‐score Test-‐1
Z-‐score Test-‐2
• Different siRNA libraries exhibit virtually no correlation. Notably, if pooling leads to cleaner and truer data, then one would expect better correlation between screens conducted with pools. This is not the case."
-‐5
-‐4
-‐3
-‐2
-‐1
0
1
2
3
4
5
-‐5 -‐4 -‐3 -‐2 -‐1 0 1 2 3 4 5 -‐5
-‐4
-‐3
-‐2
-‐1
0
1
2
3
4
5
-‐5 -‐4 -‐3 -‐2 -‐1 0 1 2 3 4 5
R = 0.03 Dharmacon Pool vs. Pool*
Z-‐score Po
ol-‐1
Z-‐score Pool-‐2
R = 0.06 Ambion Single vs. Single
Z-‐score siR
NA-‐1
Z-‐score siRNA-‐2
*All pools sharing a common sequence were removed from this analysis.
Correla+on – The Good News
If you plot siRNAs with the same seed (right) you get much beler correlaEon than if you plot different siRNAs designed to target the same gene (lem). This is clear evidence of what we already know – seed driven off target effects dominate siRNA screens. Marine, S. et al (J. Biomol. Screen, 2012)"
Off-‐Target Effects
0
50
100
150
ATCTGC TACTTC TCGTTCSeed Hexamer
Perc
ent P
ositi
ve (N
orm
alize
d to
Neg
ative
Con
trol)
TargetSCAMP5Other
Seed Type●
●
HeptamerHexamer
Biased Seeds
Screen Median
Likely False PosiEve
0
50
100
150
ACGAGA CATAGT TAGAAGSeed Hexamer
Perc
ent P
ositi
ve (N
orm
alize
d to
Neg
ative
Con
trol)
TargetXRN1Other
Seed Type●
●
HeptamerHexamer
Likely True PosiEve
Common Seed Analyis
• Common Seed Analysis, disEnguishes on vs off target effects • The visualizaEon and analysis of seed acEvity allows us to assess
if an siRNA’s acEvity is likely due to on-‐ or off-‐target effects. • This analysis can also help prioriEze sequences for downstream
studies.
Marine, S. et al (J. Biomol. Screen, 2012)"
Turning the off-‐target problem on its head
• Many siRNA strategies (pooling, raEonal design, chemical modificaEons) may miEgate off-‐target effects, but no one claims that they can be eliminated.
• Rather than explore new methods to eliminate off-‐target effects we decided to eliminate on-‐target effects.
• This may be achieved by changing bases 9-‐11, which maintains the seed sequence but eliminates on-‐target cleavage. Thus. if the “C911” seed control is sEll acEve, the original siRNA phenotype was most likely due to seed-‐driven off-‐target effects.
Buehler, E. et al (PLoS ONE, 2012)
Example Selec+ons for Gold-‐Standard True and False Posi+ves
0
50
100
150
ATAGTA GCCGTT TGTTGGSeed Hexamer
Perc
ent A
ctiv
ity TargetPOLR2AOther
Seed Type●●
HeptamerHexamer
0
50
100
150
GAGTAG TTATAG TTGTCASeed Hexamer
Perc
ent A
ctiv
ity TargetPLXDC1Other
Seed Type●●
HeptamerHexamer
Selected 10 true and 10 false posiEves from a genome-‐wide luciferase reporter screen based on Common Seed Analysis and known biology.
Buehler, E. et al (PLoS ONE, 2012)
C911 siRNAs Dis+nguish Between True and False Posi+ves
C911 modified siRNAs (inacEvated) maintain corresponding seed sequences and exhibit acEvity in the case of false posiEves (lem), but lose acEvity in the case of true posiEves (right).
False Positive True Positive
0
50
100
150
** *
***
***
** *** **
*********
******
LOC
6455
04
LHX1
LOC
6465
70
PPAP
2C
ZFP3
6L2
hCG
_198
9844
B3G
NT7
C4o
rf21
hCG
_204
5830
PLXD
C1
RPL
36
POLR
2J3
EIF2
S3
POLR
2I
PCF1
1
RPS
27A
SON
POLR
2D
POLR
2B
POLR
2A
siRNA Targets
Perc
ent L
ucife
rase
Act
ivity
siRNAunmodifiedC911
Buehler, E. et al (PLoS ONE, 2012)
0
30
60
90
120
ARR
DC
2FG
F8M
KNK1
AVIL
FBF1
ACO
T13
USM
G5
RG
S3M
FAP3
LC
HR
NA3
TMEM
25M
RPL
12LO
C64
6279
CIR
H1A
ECE2
SSR
P1R
PLP1
P5M
PHO
SPH
10C
8orf3
4C
YP4A
22BM
P4R
PL23
NLR
P6D
IO1
CAS
P7R
PL14
SLC
27A3
PATE
4R
PL3
IRX6
RTN
4RPT
PRM
RPL
35A
SZT2
EIF3
DR
PL18
AR
PL34
P34
RPL
5P1
ACTA
1EH
D2
RPS
6KA5
KLH
L7C
11or
f70
LOC
3903
64PR
SS33
RPL
30ST
X6LO
C44
2448
CAM
KVR
PLP1
NIP
AL1
CR
ABP1
ASF1
BLO
C72
9380
FSH
BSP
TLC
1AM
D1
MAN
2B2
MEF
2BN
B−M
EF2B
MZB
1C
OX1
8PU
M1
MAP
3K11
TCEA
NC
EIF3
BZI
C1
MTA
1G
FER
CH
CH
D1
XRN
1M
APK1
0LM
LND
OK5
USP
48R
HO
HEE
F2TA
P2F2
RL1
OG
FREI
F1AX
PPW
D1
MLY
CD
MIC
AH
PNF8
A2O
DC
1PT
GS2
F2R
L3R
NF5
CC
L28
EEF1
A1SL
C45
A4DA
OA
FEM
1CPO
LR3C
NO
P58
TSPA
N33
ZNF7
75PT
TG1
IGFA
LS
Gene Symbol of siRNA Target
Vira
l Spr
ead
(as
perc
ent o
f neg
ative
con
trol)
siRNAOriginalC911
Large-‐Scale Valida+on of C911
Vaccinia Viral Spread Assay Top 100 Sivan, G. et al (PNAS 2013)!
Leveraging Gene Expression
Symbol LOCUSID Descrip1on log2.fold_change p_value q_value significant IRF9 10379 interferon regulatory factor 9 1.87957 5.00E-‐05 0.000319 yes IRF2 3660 interferon regulatory factor 2 1.91502 5.00E-‐05 0.000319 yes
IRF2 (Gene ID: 3660)! IRF9 (Gene ID: 10379)!
siRNA" siRNA"
Z-Sc
ore"
Z-Sc
ore"
RNA-Seq Determination of Differential Expression +/- IFNα
RNA-Seq profiling identifies IFNα-stimulated changes in several top candidate genes that were identified through RNAi screening for modulators of IFNα-stimulated ISRE reporter activity. "
AND SEVERAL MORE…..!
Leveraging Exome Sequencing
Exome Sequencing of Drug-Resistant Line vs Parental
Exome sequencing revealed mutaEons in several top candidate genes that were idenEfied through RNAi screening for rescuing parental cells to drug amer knockdown.
Symbol LOCUSID descripEon chr_name hom_het type EIF3A 8661 eukaryoEc translaEon iniEaEon factor 3, subunit A chr10 het Resistant Line Only EPPK1 83481 epiplakin 1 chr08 het Resistant Line Only NFKB2 4791 nuclear factor of kappa light polypepEde gene enhancer in B-‐cells 2 chr10 het Resistant Line Only
Conducted a screening campaign of parental and drug-‐resistant cell lines for genes that modulate drug acEvity. Screening of the parental cells at a high dose of drug produced numerous significant candidates (leW) that resulted in rescue from drug amer knockdown. These candidates showed an enrichment for protein-‐protein interacEons (STRING) and an obvious role for the iniEaEon of translaEon.
RNAi Ac1ve
Small Molecule
Both
Leveraging Small Molecules
Parallel RNAi and focused small molecule efforts exhibit some of the same targets, connected targets, and overlapping pathways.
Gene Symbol Gene ID AURKA 6790 CHEK1 1111 HDAC1 3065 KIF11 3832 NAMPT 10135 PLK1 5347
POLR2A 5430 PSMD1 5707 RRM1 6240 RRM2 6241 TYMS 7298 WEE1 7465
Targets Iden1fied in Both RNAi and Small Molecule Efforts
Data from a simple viability screen in a rhabdomyosarcoma cell line.
Screening for Novel Drug Combina+ons
• Increased efficacy • Delay resistance • Alenuate toxicity
• Inform signaling pathway connecEvity
• IdenEfy syntheEc lethality • Highlight polypharmacology
Transla+onal Interest Basic Interest
Screening for Novel Drug Combina+ons
• Lots of ways to predict synergisEc combinaEons – Li et al (BioinformaEcs, 2014) is a comprehensive approach
• Efforts to link combinaEon screens to paEents have also started (Crystal et al, Science, 2014)
• Robust predicEons require mulEple data types – Small molecule acEvity (on-‐ and off-‐target) – Gene expression (and other genomic data) – Biological connecEvity
Mechanism Interroga+on PlateE • CollecEon of ~ 2000 small molecules of diverse mechanism of acEon. • 745 approved drugs • 420 phase I-‐III invesEgaEonal drugs • 767 preclinical molecules
• Diverse and redundant MOAs represented
AMG-47a Lck inhibitor Preclinical
belinostat HDAC inhibitor Phase II
Eliprodil NMDA antagonist Phase III
JNJ-38877605 HGFR inhibitor Phase I
JZL-184 MAGL inhibitor Preclinical
GSK-1995010 FAS inhibitor Preclinical
Combina+on Screening Workflow
Run single agent dose responses
6x6 matrices for poten'al synergies
10x10 for confirma'on + self-‐cross
Acoustic dispense, 15 min for 1260 wells, 14 min for
1200 wells"
Characterizing Synergy
• Many models have been devised to describe the response of two drugs when combined – Highest single agent (aka Gaddum) – Loewe – Bliss
• Based on these we can calculated a variety of metrics that will indicate whether two drugs exhibit a synergisEc, addiEve or antagonisEc response
Analysing Combina+ons in Aggregate
• ComputaEons on individual combinaEons are useful – IdenEfy promising candidates, prioriEze for followup
• How can we examine combinaEons in aggregate? • What can we find if we look at combinaEons in aggregate? – Similar combinaEons – CombinaEon behavior across cell lines
Network Representa+ons
CombinaEon screens lend themselves naturally to network representaEons
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●● ●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
● ●●
●
●●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
∆ Bliss+
−4.3
−3.8
−3.3
−2.9
−2.4
−1.9
−1.4
−1.0
−0.5
0.0
●
●
●
●
●
●
●
●
●●
●
●
●
● ●
●●●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
∆ Bliss+
−3.4−3.1
−2.7
−2.3
−1.9
−1.5−1.2
−0.8
−0.4
0.0
immune system process
apoptotic process
transcription from RNApolymerase II promoter
protein phosphorylation
cell communication
immune response
Network Representa+ons
• Things get more interesEng when we have n m screens
• Can be simplified using a variety of methods – Neighborhoods – Minimum Spanning Tree
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
×
Comparing Neighborhoods
CombinaEons that have DBSumNeg < 1st quarEle value for that strain
3D7 DD2 HB3
Comparing Neighborhoods
AlternaEvely, consider all tested combinaEons, highlighEng distribuEon of synergisEc and antagonisEc combinaEons
3D7 DD2 HB3
Iden+fying the Most Synergis+c Pairs
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
Cluster C4
• Focus on sugar metabolism
• Ruboxistaurin, cycloheximide, 2-‐methoxyestradiol, …
• PI3K/Akt/mTOR signalling pathways glycogen metabolic process
regulation of glycogen biosynthetic process
glucan biosynthetic process
glucan metabolic process
cellular polysaccharide metabolic process
regulation of generation of precursor metabolites and energy
peptidyl-serine phosphorylation
cellular macromolecule localization
regulation of polysaccharide biosynthetic process
cellular carbohydrate biosynthetic process
0 1 2 3-log10(Pvalue)
0.00
0.02
0.04
0.06
0.08
361
254
215
164
143 82 125
327
241
194
145
116
139
371
163
165
384
339
322
217
184
150 52 136
Combina+ons across Cell Lines
• Cellular background affects responses • Can we group cell lines based on combinaEon response?
• Or find “fingerprints” that characterize cell lines?
Many Choices to Make 0
12
34
KMS-34
INA-6
L363
OPM-1
XG-2
FR4
AMO-1
XG-6
MOLP-8
ANBL-6
KMS-20
XG-7
OCI-MY1
XG-1
8226
EJM
U266
KMS-11LB
SKMM-1
MM-MM1
sum
0.0
0.1
0.2
0.3
0.4
0.5
0.6
L363
OPM-1
XG-2
KMS-20
XG-1
XG-7
ANBL-6
OCI-MY1
U266
XG-6
INA-6
MOLP-8
AMO-1
KMS-34
KMS-11LB
SKMM-1
MM-MM1
EJM FR4
8226
max
0.00
0.05
0.10
0.15
0.20
0.25
INA-6
MM-MM1
8226
XG-1
U266
ANBL-6
SKMM-1
EJM
OPM-1
XG-2
OCI-MY1
KMS-20
L363
KMS-11LB
AMO-1
XG-6
FR4
KMS-34
MOLP-8
XG-7
min
0.0
0.2
0.4
0.6
0.8
1.0
1.2
L363
OPM-1
XG-2
KMS-34
INA-6
KMS-11LB
SKMM-1
EJM
U266
MM-MM1
FR4
AMO-1
XG-6
8226
MOLP-8
ANBL-6
OCI-MY1
XG-1
KMS-20
XG-7
euc
• Vargatef exhibited anomalous matrix response compared to other VEGFR inhibitors
Exploi+ng Polypharmacology
Vargatef
Linifanib Axitinib Sorafenib Vatalanib
Motesanib Tivozanib Brivanib Telatinib
Cabozantinib Cediranib BMS-794833 Lenvatinib
OSI-632 Foretinib Regorafenib
Exploi+ng Polypharmacology
• PD-‐166285 is a SRC & FGFR inhibitor
• Lestaurnib has acEvity against FLT3
Vargatef DCC-2036 PD-166285 GDC-0941
PI-103 GDC-0980 Bardoxolone methyl AT-7519AT7519
SNS-032 NCGC00188382-01 Lestaurtinib CNF-2024
ISOX Belinostat PF-477736 AZD-7762
Chk1 IC50 = 105 nM
VEGFR-1
VEGFR-2
VEGFR-3
FGFR-1
FGFR-2
FGFR-3
FGFR-4
PDGFRa
PDGFRb
Flt-3
Lck
Lyn
Src
0 200 400 600Potency (nM)
Hilberg, F. et al, Cancer Res., 2008, 68, 4774-‐4782