PIONEERING NEW DISCOVERIES IN HUMAN DISEASE BIOLOGY. THINK HUMAN. HEMOSHEAR NONCONFIDENTIAL In Vitro Liver Models And Their Applications For NASH 30 th June 2016 Ajit Dash, MD, PhD Senior Scientific Director HemoShear Therapeutics
PIONEERING NEW DISCOVERIES
IN HUMAN DISEASE BIOLOGY.
THINK HUMAN.
HEMOSHEAR NONCONFIDENTIAL
In Vitro Liver Models And Their Applications For NASH
30th June 2016
Ajit Dash, MD, PhD Senior Scientific Director
HemoShear Therapeutics
HEMOSHEAR NONCONFIDENTIAL
Non Alcoholic Steatohepatitis: Disease Problem and Unmet Needs
An ideal in vitro liver
model would fulfill various
unmet needs:
Unbiased novel target
discovery
Development of non-
invasive translational
biomarkers for diagnosis
and monitoring disease.
Understanding/predicting
efficacy differences, in
stratified sub-populations
(Personalized medicine).
Safety assessment under
disease-like conditions.
Target 2
Target 1
Target 3
Target 4
Source: Catenion analysis based on ADIS database,
2014
1 in 3 adults in the U.S. has non-alcoholic fatty liver
disease
75% of people with NASH also have type 2 diabetes
Fastest growing disease in China and India.
Approximately 50 active programs with 38 distinct
therapeutic targets
HEMOSHEAR NONCONFIDENTIAL
Underlying Mechanisms of Steatohepatitis are Complex
1 2
3
4
Fatty acid
β-oxidation ( )
Peripheral fatty
acid mobilization
and hepatic
uptake ( )
Lipoprotein
export ( )
PXR
LXR
PPARG
ChREBP
SREBP-1c
MTP activity
VLDL
Secretion
De novo
Fatty acid and
triglyceride
synthesis ( )
CD36
FASN
ACC
SCD1
ELOV
CPT-1
VLCAD
LCAD
SCAD PGC1A
PPARA
SIRT1
AMPK
+ -
+ -
FABP
Lipid
Peroxidation and
Oxidative stress
(ROS) ( )
Peroxisomal
FA oxidation,
CYP2E1
Cytokines (TFGβ, IL-8,
TNFα)
Inflammatory
Cell
Recruitment
Stellate Cell
Activation and
Extracellular
Matrix
Metabolic
Imbalance
Drugs
Viruses
Mechanisms of
Steatosis
1. Synthesis of
lipids/cholesterol
2. β-Oxidation
of fatty acids
3. Export of
lipoproteins
4. Uptake of
fatty acids
Steatosis Oxidative
Stress
Inflammatory
Cytokines
Macrophage/St
ellate Activation
Extracellular
Matrix
Deposition
HEMOSHEAR NONCONFIDENTIAL
Existing in vitro Models: Challenges and Opportunities
Species Cell Type(s) Origin
Human
Hepatocytes Primary
(Healthy/Patient)
Huh7 Hepatoma
HepG2 Hepatocellular Ca
Hepatic Stellate Cells Healthy/Patient
LX2 Stellate Cell line Immortalized
Hepatocytes +
Adipose Cells
Huh7 + LX2
Canine Hepatocytes Primary
Rat
Primary Hepatocytes
H4IE Immortalized
H4IEC3 Immortalized
PAV-1 Immortalized
Mouse
RAW 264.7
Macrophages and
AML-12 Cell co-
cultures
Immortalized
Challenges of existing models employing
static flat-plate cell cultures:
Dedifferentiation and loss of CYP activity.
Non-physiological levels of glucose and
insulin and loss of insulin sensitivity.
Altered baseline inflammatory state.
Hypoxia-reperfusion on media change.
Non-relevant drug and metabolite
concentration profiles
Opportunities for improvement:
Organotypic approaches (3-D, heterotypic
cell interactions, flow).
Physiological media formulations and drug
concentrations based on clinical
pharmacokinetics
Use of Translational biomarkers.
Big data –omic approaches
HEMOSHEAR NONCONFIDENTIAL
Recreating Physiological Milieu and Parameters in a 3D Culture
Configuration
3D cell configuration - modeled on sinusoid
with hepatocytes ± non-parenchymal cells.
Simultaneous perfusion and hemodynamics -
allows control of drug, nutrient and oxygen
gradients
Effluent and cells can be assessed from top
and bottom separately.
Hemodynamic Flow Transmural Flow
Interstitial Flow
Transport Inflow
Transport Outflow
Hemodynamic Flow
Interstitial Flow
Adapted from: Nature Reviews Immunology 14, 181–194 (2014)
Dash et al, AJP 2013, Terelius et al Chem Bio 2015, Chapman et al 2016
Hepatocytes Plated
1. RNA-Seq Analysis
2. Functional Endpoints e.g. MTT, Imaging, CYP Assays
3. Secreted Biomarkers e.g. Albumin, Cytokines, FGF19
Treatment
(3- 7 days) (2-7 days)
Restoration of Biology
Day 14 Controlled Hemodynamics: In vivo liver:
MRP-2, HNF-4α
Day 14 Controlled Hemodynamics: In vivo liver:
MRP-2, HNF-4α
HEMOSHEAR NONCONFIDENTIAL
Liver-like Polarized Morphology and Function Maintained Over Time
Dash et al SOT 2013, Marukian et al AASLD 2013
6
In vivo Liver
E-cadherin HNF4α CD26, Draq5 MRP2 HNF4α CD81, Draq5
0
5
10
15
20
25
30
Day 2 Day 4 Day 7 Day 9
HemoShear
Static
0
100
200
300
400
500
600
700
Day 2 Day 4 Day 7 Day 9
HemoShear
Static
0
2
4
6
8
10
12
CYP1A2 CYP2B6 CYP3A4/5 CYP2C9 CYP2D6
Controlled Hemodynamics
Static
CYP Activity
10#μm#
Transporter Activity CYTOCHROME P450
ACTIVITY
CDFDA -> CDF
TRANSPORTER
ACTIVITY
ALBUMIN SECRETION
(ANABOLIC)
UREA SECRETION
(CATABOLIC)
TIGHT JUNCTIONAL
PROTEIN
SURFACE ANTIGENIC
ENZYME
BILIARY EFFLUX
TRANSPORTER
SURFACE
GLYCOPROTEIN
HEMOSHEAR NONCONFIDENTIAL
Drug Responses Exhibited at Clinically Relevant Concentrations
7
Figler et al AASLD 2015.
Rumack-Matthews nomogram for serum concentration thresholds for clinical treatment of Acetaminophen poisoning.
B
C
Toxic in HemoShear System
A Safe in HemoShear System
Toxicity seen in Other Systems
FDA guidelines
for antidote
administration.
Normal ‘safe’
concentration
Eff
ica
cy
/To
xic
ity
Log Scale Concentration
In vivo/Clinical Toxicity
In vivo/Clinical Efficacy
Therapeutic Window
Cmax 10Cmax
Toxicity
Subclinical Efficacy
100Cma
x
In vitro Toxicity
In vitro Efficacy
Exp
res
sio
n/A
cti
vit
y L
ev
el
Time (Days)
Localized Transporter Activity In vitro
Phase I and II enzyme levels/activity in Static
Phase I and Phase II enzyme levels/activity
under controlled hemodynamics
5 10
A
B
Efficacy and toxicity responses seen at
concentrations that match clinical
therapeutic exposures.
Over 30 drugs assessed for mechanistic
differences using transcriptomics. (NIH
SBIR Award R44 DK091104-02)
Dash et al Expert Opinion 2012
HEMOSHEAR NONCONFIDENTIAL Deering et al AASLD, 2012, Dash et al ADA 2013, Cole et al AASLD 2015
0
20
40
60
80
100
120
140
160
Healthy Disease Disease + Pioglitazone 1.5 uM Healthy Diabetes Diabetes
+ Pio
0
10
20
30
40
50
60
70
80
90
100
Healthy + DMSO Steatotic + DMSO Steatotic + Pioglitazone 1.5 uM Healthy Diabetes Diabetes
+ Pio
Gluconeogenesis
Different ially Regulated CYP Transcripts
HG/HI LG/LI ‘Diabet ic’ MilieuHealthy Milieu
CYPs
involved in
drug
metabolism
CYPs
involved in
cholesterol/
lipid
synthesis
ug/m
g o
f pro
tein
F
old
over
sta
tic
8
Insulin Sensitivity And Lipogenic Responses Maintained
Actin Lipid (Nile Red) Nucleus
Normal Glucose/Insulin
(Healthy Milieu) High Glucose/Insulin
(Steatotic Milieu)
Days in-device (or in static cultur e)
0
5
10
15
20
25
day2 day4 day7 day9 day10
ng
/day/1
0^
6c
ells
Static HG-HI
Device LI-LG
Static conditions: High insulin/glucose
HemoShear: Low insulin/glucose
HS System
Static Culture
0
0.5
1
1.5
2
2.5
noRx Ins LP LP+In
Medium 1
ng
/mg
pro
tein
Glycogenolysis
Gluconeogenesis Glycogenolysis +
Gluconeogenesis
Lactate/
Pyruvate
Basal Insulin Lactate/
Pyruvate +
Insulin
LIPOPROTEIN (a) SECRETION
GLUCONEOGENESIS DE NOVO LIPOGENESIS
TRIGLYCERIDES
CYP3A2
CYP GENES
Healthy Steatotic
Healthy Steatotic Healthy Steatotic
Insulin sensitivity allows culture in a close to physiologic milieu and altered disease-like
steatotic phenotype under hyperglycemic, hyperinsulinemic conditions.
HEMOSHEAR NONCONFIDENTIAL
Applications of a Physiologically Responsive Liver Model
9
After demonstrating that the system maintained differentiated liver phenotype
as evidenced by polarized morphology, liver specific functions, drug
metabolizing enzyme and transporter activity and responsiveness to insulin,
we tested the model for the following applications:
1. Assessing on-target and off-target pharmaco-toxicology of drugs at
clinically relevant concentrations.
2. Distinguishing transcriptomic signatures of various phenotypes of drug
induced liver injury (DILI).
3. Studying underlying mechanisms of drug induced steatohepatitis that
could help understand potential NASH targets.
4. Developing a lipotoxic model with milieu mimicking metabolic disease.
HEMOSHEAR NONCONFIDENTIAL
Assessing On-target Pharmacology of Obeticholic Acid
10
18
74
8
comparison
Responses (NA)
CYP7A1
SLC27A5
CYP27A1
AKR1D1
HSD3B7
ACOT8
AKR1C4
BAAT
CYP8B1
AMACR
SCP2
ACOX2
HSD17B4
SLC27A2
ABCB11
Gene
s (
0.9
8)
. .
. .
. .
.
.
.
. .
.
. .
. .
. .
Gene Response Heatmap
−2 −1 0 1 2
Value
02
4
Co
un
t
16
44
3
comparison
Responses (NA)
CYP7A1
SLC51B
SLC51A
NR0B2
ABCB11
ABCB4
RXRA
SLCO1B1
ADCY3
ADCY8
SLCO1B3
ADCY1
CFTR
PRKX
NR1H4
ABCC2
GNAS
SLC10A1
ADCY6
ADCY9
PRKACA
PRKACB
SCTR
ADCY5
Gene
s (0.96
)
. .
. .
. .
. .
. .
. .
. .
. .
.
.
.
. .
.
.
.
Gene Response Heatmap
−2 −1 0 1 2
Value
04
8
Co
unt
Bile
Syn
the
sis
B
ile S
ecre
tio
n
18
74
8
comparison
Responses (NA)
CYP7A1
SLC27A5
CYP27A1
AKR1D1
HSD3B7
ACOT8
AKR1C4
BAAT
CYP8B1
AMACR
SCP2
ACOX2
HSD17B4
SLC27A2
ABCB11
Gene
s (
0.9
8)
. .
. .
. .
.
.
.
. .
.
. .
. .
. .
Gene Response Heatmap
−2 −1 0 1 2
Value
02
4
Co
un
t
18
74
8
comparison
Responses (NA)
CYP7A1
SLC27A5
CYP27A1
AKR1D1
HSD3B7
ACOT8
AKR1C4
BAAT
CYP8B1
AMACR
SCP2
ACOX2
HSD17B4
SLC27A2
ABCB11
Gene
s (
0.9
8)
. .
. .
. .
.
.
.
. .
.
. .
. .
. .
Gene Response Heatmap
−2 −1 0 1 2
Value
02
4
Co
un
t
16
44
3
comparison
Responses (NA)
CYP7A1
SLC51B
SLC51A
NR0B2
ABCB11
ABCB4
RXRA
SLCO1B1
ADCY3
ADCY8
SLCO1B3
ADCY1
CFTR
PRKX
NR1H4
ABCC2
GNAS
SLC10A1
ADCY6
ADCY9
PRKACA
PRKACB
SCTR
ADCY5
Gene
s (0.96
)
. .
. .
. .
. .
. .
. .
. .
. .
.
.
.
. .
.
.
.
Gene Response Heatmap
−2 −1 0 1 2
Value
04
8
Co
unt
16
44
3
comparison
Responses (NA)
CYP7A1
SLC51B
SLC51A
NR0B2
ABCB11
ABCB4
RXRA
SLCO1B1
ADCY3
ADCY8
SLCO1B3
ADCY1
CFTR
PRKX
NR1H4
ABCC2
GNAS
SLC10A1
ADCY6
ADCY9
PRKACA
PRKACB
SCTR
ADCY5
Gene
s (
0.9
6)
. .
. .
. .
. .
. .
. .
. .
. .
.
.
.
. .
.
.
.
Gene Response Heatmap
−2 −1 0 1 2
Value
04
8
Co
un
t
10 μM
Obeticholic Acid
0.5 μM comparison
Responses (NA)
FGFR4
NR5A2
NR1H4
NR0B2
FGF19
Gene
s (0
.98
)
. .
. .
.
. .
. .
Gene Response Heatmap
−2 −1 0 1 2
Value
02
Co
unt
FX
R S
ign
alin
g
100
1000
DMSO INT_LOW INT_HIGH
TreatmentC
once
ntra
tion Donor
HC5−30
Hu8160
QHu0032
TRL4038
FGF19Crossbars: 95% Confidence Inter val based on model
Secreted FGF19 Protein
OCA 0.5μM 0.5μM 10μM 0.1%
Obeticholic Acid DMSO
***
*** p<0.001
***
***
Concentr
ation (
pg/m
l)
Donor
2
5
1
3
1000
100
SHP
NTCP
FXR-RXR
FXR
CYP7A1 HMGCR
Cholesterol Synthesis
Bile Acid Synthesis
FXR
BSEP
Bile Acid Secretion
ABCB4
CYP27A1
FGF19
Obeticholic Acid
Relative to DMSO Controls
-2 2
LOG2 Fold Change
FGF19
CYP7A1
Strongly induced
FGF19 in hepatocytes,
both at a gene and
protein level, confirming
a direct hepatic effect in
addition to the widely
appreciated FGF19
loop through the gut.
CYP7A1 was the most
down-regulated
differentially expressed
gene in the
transcriptome, with
simultaneous down-
regulation of the bile
synthesis pathway
genes.
Sanyal, Oral presentation AASLD 2015
HEMOSHEAR NONCONFIDENTIAL
Pharmaco-toxicological Signature of Obeticholic Acid and Impact of CYP
Activity
11
0,0
0,5
1,0
1,5
2,0
Bile Secretion (*,#)
Bile Synthesis (*)
FXR GeneTranscription (*,#)
CholeseterolBiosynthesis (*)
TriglycerideSynthesis (*,#)
IL6 Signaling (*)
Reactive OxygenSpecies (*)
Apoptosis
ExtracellularMatrix
Vehicle Control
0.5uM Obeticolic Acid
10uM Obeticolic Acid
1
10
DMSO OCA_0.5_uM OCA_10_uM
Treatment
Con
cent
ratio
n man_don
1
2
3
5
CYP3a4Crossbars: 95% CI based on linear model
Error bars: +/− SE
1
10
DMSO OCA_0.5_uM OCA_10_uM
Treatment
Con
cent
ratio
n man_don
1
2
3
5
CYP1a2Crossbars: 95% CI based on linear model
Error bars: +/− SE
OCA 0.5μM OCA 10μM
OCA 0.5μM OCA 10μM
0.5μM 10μM 0.1%
Obeticholic Acid DMSO
0.5μM 10μM 0.1%
Obeticholic Acid DMSO
Concentr
ation
(nM
/μg p
rote
in)
Donor 1 2 3 5
Donor 1 2 3 5
**
***
**
** p<0.01
*** p<0.001
10
1
Concentr
ation
(nM
/μg p
rote
in)
10
1
CYP3A4 Activity
(Hydroxy-Testosterone Formation)
CYP1A2 Activity
(Acetaminophen Formation)
Pathway analysis and scoring confirmed beneficial effects
of obeticholic acid on reducing steatotic indices and
inflammatory signaling.
Functional CYP assays revealed that obeticholic acid
suppressed CYP1A2 and CYP3A4 activity.
HEMOSHEAR NONCONFIDENTIAL
Distinguishing Drug Induced Steatohepatitis Signatures From Other Forms
of Drug Induced Liver Injury
12
18
405
18
559
comparison
Responses (0.83)
GPD1LDGAT2LPIN3GPD1FDPSELOVL6FDFT1DHCR7HSD17B7MVDCYP51A1ACLYLPCAT1MVKNSDHLMSMO1SQLESC5DHMGCRIDI1ELOVL1HSD17B12TM7SF2HMGCS1EBPLSSLPIN1AGPAT5AGPAT6ACSL1
Gene
s (0.87
)
. .. .
. .. . .
. . .
. . . .
. .
. .
. . .
. . .
. . .
. .
. .
. .
. . .
. .
. . . .
. .. .
. .
. .
. .
. . .
. . .
. .
. .. .
.. . . .
. .
Gene Response Heatmap
−2 −1 0 1 2
Value
06
Co
unt
APAP RIT AMI TVX
0,00
1,00
2,00
3,00
4,00
5,00
6,00
Vehicle Amiodarone
Relative to DMSO Controls
-2 2
LOG2 Fold Change HSB-101
(Steatoinflammatory biomarker)
Figler et al AASLD 2015.
Transcriptomic analysis allowed us to characterize distinct
signatures for different drugs having different DILI phenotypes.
Assays like Nile Red (neutral lipid) and secreted protein
biomarkers in effluent media were confirmatory functional
endpoints that defined the steatohepatitic phenotype.
0
50
100
150
200
Lipid &Choleste
rolSynth…
TCAcycleand
Respir…
Oxidative Stress
Glutathione
Detoxification
Coagulation
NFkBPathwayActivatio
n
TNF-aStress
Signaling
TGF-BSignalin
g
Unfolded
ProteinRespo…
Bilemetabolism andsecreti…
Trovafloxacin
Acetaminophen
Amiodarone
Ritonavir
Vehicle
NFkB-
Signaling
Coagulation
Glutathione
Detoxification
Oxidative
Stress
TCA Cycle
and
Respiration
Lipid and
Cholesterol
Synthesis
Bile
Metabolism
and
Secretion
Unfolded
Protein
Response
TGF-B
Signaling
TNF-A
Stress
Signaling
Phalloidin, Nile Red, ToPro
Vehicle
Obeticholic Acid 10µM
Ketoconazole 65.9µM
Amiodarone 39µM
HEMOSHEAR NONCONFIDENTIAL
Understanding Mechanisms and Potential Targets of Steatohepatitis
13
0
1
2
3
4
5
6
Vehicle Amiodarone Ketoconazole NASH-like milieu
1699
2
comparison
Responses (NA)
MAP2K6CDK1CD36DUSP6PPP2R1BRPS6KA2JUNPPP2CBFOSRPS6KA5SIGIRRTYK2MAP2K7NOD1MAPK7MAPK11IRAK1ECSITIRAK2CD14IL6RNOD2PELI3DUSP4LY96MAPK3TLR2PELI2MAPKAPK3RPS6KA1
Gene
s (0.7
6)
. .
.
. .
. . .
. . .
. .
. . ..
. . ..
. . ...
.
..
. . .
. . .
. . .
. .
. .
. . .
. .
. .
. .
. .
. .
. .
. .
. .
Gene Response Heatmap
−2 −1 0 1 2
Value
04
Co
un
t
0
50
100
150
200
TriglycerideSynthesis
Fatty Acid Oxidation
Lipid Transport
MitochondrialDysfunction
Respiration and ATPSynthesis
Bile Secretion
myd88 Signaling
NRf2 signaling
IL6 Signaling
TNF Signaling
AMI
KET
OCA
Vehicle
2430
0
comparison
Responses (NA)
GPD1
DGAT2
AGPAT1
ACSL6
ELOVL2
ACSL4
FASN
ELOVL6
ELOVL7
GPAM
AGPAT2
ACSL1
ELOVL1
ACSL5
ACACA
TECR
SLC25A1
LPCAT4
ELOVL5
AGPAT5
LPIN2
AGPAT9
ACSL3
AGPAT6
LPCAT1
LPIN1
HSD17B12
ACLY
Gen
es (0
.8
)
. . .
. . .
. . ..
. . .. .. .
. . .
. . ..
. .. . .. . .. . .. . .
. .
. .
.. . .. .. .. .. .. ... . .. .. .
Gene Response Heatmap
−2 −1 0 1 2
Value
04
Co
unt
TR
IGL
YC
ER
IDE
SY
NT
H
MY
D8
8 S
IGN
AL
ING
1712
0
comparison
Responses (NA)
NDUFS7NDUFB7ATP5DNDUFV1COX8ANDUFA3UQCR10UQCR11NDUFS8CYC1NDUFA6NDUFA11NDUFV2ATP5C1SDHCNDUFB4ETFDHNDUFAB1ATP5J2UQCRHUQCRFS1NDUFS3NDUFS5NDUFS2ETFACOX5ACYCSNDUFA8COX7A2LUCP2
Gene
s (0.91
)
.
.. .
.
.
.
.. .. .
. .
. . .. .
. . .... .. .
. .
. .
. .
. .
. ....
.
. .
. . .
. . ..
Gene Response Heatmap
−2 −1 0 1 2
Value
06
Co
unt
comparison
Responses (NA)
TRAF2RPS6KA4MAP2K7MAP3K14AKT1MAP3K5MLKLMAPK11AKT2MAPK12RPS6KA5CREB3L1PIK3CDPIK3R1MAP2K6BIRC3CREB5CREB3L4NFKB1MAP3K8CASP10JUNTNFRSF1BMAGI2ATF4PIK3R3TRAF5RIPK3MAPK3CREB3L3
Gene
s (0.77
)
.
.
.. .. .
.. .
.. . .
.
.. .
. .
. . .
. .
. . .
. .
. . .
. .
. . ..
. . .
. ..
. .
. .
. .
. .
. .
. . .
Gene Response Heatmap
−2 −1 0 1 2
Value
04
Co
unt
RE
SP
CH
AIN
/ A
TP
TN
F-α
SIG
NA
LIN
G
Relative to DMSO Controls
-2 2
LOG2 Fold Change
AMI KET OCA AMI KET OCA
AMI KET OCA AMI KET OCA
AMI KET Healthy Lipotoxic
Media
HSB-101
(Steatoinflammatory biomarker)
Differential analysis of transcriptomic signatures for different drugs causing steatohepatitis
versus those causing NASH may offer insights into mechanisms and targets.
Dash et al EASL 2016.
HEMOSHEAR NONCONFIDENTIAL
Liver Models to Recapitulate Metabolic Disease Spectrum
14
0,5
1
2
4
HEALTHY NAFLD NASH NASH+TNF
Rela
tive N
ile R
ed (
FO
LD
) INFLAMMATORY CYTOKINES LIPID ACCUMULATION
2
8
32
128
CX
CL
10
(p
g/m
l)
1
1
2
4
Healthy NAFLD NASH
Rela
tive T
rigly
cerides
(FO
LD
)
HEPATIC TRIGLYCERIDES
FBP1 FBP1 -
-
-
-
+
+
-
-
+
+
+
-
+
+
+
+
-
-
-
-
+
+
-
-
+
+
+
-
+
+
+
+
FBP1 FBP1 GLUCOSE:
INSULIN:
FFA:
-
-
-
+
+
-
+
+
+
GLUCOSE:
INSULIN:
FFA:
TNF:
GLUCOSE:
INSULIN:
FFA:
TNF:
HEALTHY FATTY
LIVER
FATTY LIVER +
INFLAMMATION
ADVANCED
LIPOTOXIC
Healthy Milieu Steatotic Milieu
(Glucose, Insulin)
Lipotoxic Milieu
(Steatotic + Fatty
Acids)
Lipotoxic Milieu +
TNF
Lipotoxic metabolic disease model has Kupffer cells and stellate cells added on opposite side of
the membrane
HEMOSHEAR NONCONFIDENTIAL
Ongoing Validation of Drug Responses in the Advanced Lipotoxic Liver
System
In the advanced lipotoxic liver system, OCA
Reduced ALT levels (a clinical biomarker for
NASH)
Promoted a robust increase in downstream
targets of FXR signaling, including FGF19
Reduced several markers of inflammation
Reduced markers of fibrosis
ON-TARGET FXR
SIGNALING
0,5
1
VEH VEH OCA
Healthy NASH
ALT
(F
OLD
)
CLINICAL
BIOMARKERS p=0.12
**
**p<0.005,* p<0.05 relative to VEH
**
0,25
0,5
1
IL8 CHI3L1 IL6 MCP1 GRO VEGF
Secre
ted P
rote
in
Concentr
ations (
FO
LD
)
INFLAMMATORY BIOMARKER
PANEL
*
p=
0.1
3
** **
*
0,5
1
2
4
8
16
32
VEH OCA
NASH
FG
F19 (
FO
LD
)
0,125
0,25
0,5
1
Secre
ted P
rote
in C
oncentr
ations
(FO
LD
)
FIBROSIS
BIOMARKER PANEL
TGFβ OPN Pro-
Col
*
-100%
-50%
0%
50%
100%
150% Cholesterol Biosynthesis
Fatty Acid Synthesis
Cytokine-cytokine receptor interaction
TGFB Signaling
Glucose Regulation
-100%
-50%
0%
50%
100%
150% Cholesterol Biosynthesis
Fatty Acid Synthesis
Cytokine-cytokine receptor interaction
TGFB Signaling
Glucose Regulation
NASH
Diabetic
Healthy
OCA
LY2109761
15
Advanced
Lipotoxic Advanced
Lipotoxic Advanced Lipotoxic Advanced Lipotoxic
HEMOSHEAR NONCONFIDENTIAL
Conclusions and Future Directions
16
• The physiologically responsive liver model allows assessment of on-
target and off-target pharmaco-toxicology of drugs at clinically relevant
concentrations.
• Comparative analysis of transcriptomic responses of drug induced
steatohepatitis, lipotoxic NASH-like conditions and drugs that impact
NASH could help identify and understand potential NASH targets.
• Ongoing/Future activities include:
• Benchmarking signatures against clinical samples.
• Analysis of non-parenchymal cell response within system.
• Lipidomic analysis to gain a better understanding of lipid fractions
under lipotoxic milieu and how they correlate with transcriptomics.
• Characterization of translatable functional responses such as
histology and extracellular matrix composition measurements.
• Comparative analysis of drug response under healthy versus
lipotoxic conditions and stratified patient derived hepatocytes
versus human hepatocytes could provide additional insights about
useful applications of this system.
HEMOSHEAR NONCONFIDENTIAL
Acknowledgements and Funding
Entire HemoShear Scientific Team
17
Brian Wamhoff, PhD Head of Innovation
Robert Figler PhD Senior Director
Ryan Feaver, PhD Program Leader, NASH
Banu Cole, PhD Director
NIH SBIR Grants:
• R43/R44 DK100136:
Development of an in vitro system
of human hepatic steatosis.
• R44 DK104456-01: Development
of a human physiological multi-
cellular liver platform for drug-
induced liver injury and disease.
• R43/R44DK091104: Development
of a human hepatocyte predictive
pharmacology and toxicology
system.
• R44GM109539: Development of
an iPSC-derived human hepatocyte
platform for drug development.
Arun Sanyal, MD Consultant & Collaborator