A Toolbox and Blueprint for Comparability Testing in Process

Development

Robert Deans, Athersys Inc

ISCT NA Regional Meeting

9.9.13

Where is Cell Comparability Critical in the Development Path?

• Analytical tools support consistency in product manufacturing and process change towards commercialization

• Donor comparability is key whether comparing master cell banks or large donor sets

• Large scale manufacturing moves towards extended population doublings and need to assess safety and potency

• Ensuring distinction between competing products in adherent stem cell space – enforcing and growing IP boundaries

• Building a Regulatory strategy for Implementing major process changes (xeno-free media; alternate bioreactor work; driving down COGS)

Critical Product Attributes

• Potency demonstrated in pre-clinical disease model with hypothesis for mechanistic pathway supported by in vitro surrogates– Angiogenesis in vivo linked to expression of angiogenic factors

• Cell size, stability effective with route of delivery and devices

• Cytogenetic stability

• Replicative capacity and scalability for manufacturing capable of meeting commercial needs

• Phenotype and characterization meeting intellectual property definitions and distinction from competitors

• Cautious eye towards biological properties (eg., tissue migration) that may not manifest ex vivo

Cell Equivalency Under Current Process

Cell equivalency assays

Growth kinetics – telomerase activity

Viability, attachment post thaw

Defined cell population in size, granularity

Acceptable expression of CXCL5, VEGF & IL8

Valid flow cytometric profile - identity and purity

Normal karyotype

qPCR marker expression

Acceptable differentiation to osteo, adipo, chondro

Activity in T cell proliferation assay

Activity in angiogenesis assay Automated quantitation of vascular tube formation

assay

Advantaged Expansion Profile for MultiStem vs. MSC

5

Cell Expansion over TimeHuman Cells Isolated From Same Donor

Doublings

Days

MultiStem (40x)~15-20 microns

MSC (40x)~25-30+ microns

*MultiStem Different in Size than MSC and other Bone Marrow MNCs

MSC (red)

MultiStem (green)

Lymphocytes

Monocytes

6

MSC

Lew

is

pretrea

ted M

SC L

ewis

Multi

Stem

0

10

20

30

ce

ll s

ize

µm

Flow Cytometric Analysis Of HumanMultiStem By Forward / Side Scatter Plot

Microscopic Size Comparisonof Rat MultiStem and MSC

AVAILABLE AS ATHERSYS STUDY REPORT

CONFIDENTIAL

• Telomerase activity is preserved in XF-MS

7

Telomerase activity

y i o y i o y i o0

1.0107

2.0107

3.0107

MSC MultiStem Xeno-freeMultiStem

Te

lom

era

se

activi

ty (

co

py n

°)

MultiStem and MSC have Different Angiogenic Factor Profiles

MSC147 P4

MultiStem147 P4

MSC149 P6

MultiStem149 P6

MSC153 P6

MultiStem153 P6

• Angiogenic Factor Immunoblot

– Capture Assay using 60 angiogenic factor mAB

• Comparison of activity from MSC (top) and MultiStem(bottom)

– MSC / MultiStem from same donor bone marrow

– Repeated across three donors

8

Distinctive Transcription Profiles for MultiStem and MSC – Can Be Applied to Distinguish HCT/P Products

9

Blinded analysis of gene expression by an independent lab demonstrates that:(1) MSC gene expression is tightly correlated with other MSC samples;(2) MultiStem gene expression is tightly correlated with other MultiStem samples; and(3) MSC and MultiStem gene expression patterns are distinctive.

X-axis represents the correlation value.

0.125 0.100 0.075 0.050 0.025 0

ATH113

ATH114

ATH102

ATH104

ATH107

ATH103

ATH106

ATH105

ATH117

ATH116

ATH115

ATH110

ATH111

ATH112

MultiStem

MSC

AVAILABLE AS ATHERSYS STUDY REPORT

Distinctive Protein Expression – Representative Examples

10

MultiStem MSC

MultiStem MSC

Caveolin

Prostacyclin Synthase

Target Marker Detection using FITC (green) FluoresenceBlue Nuclear Counterstain, Red Actin Filament Counterstain

AVAILABLE AS ATHERSYS STUDY REPORT

Scale-Up vs Scale-Out Manufacturing Options

• 40 Layer Cell Factory, Hyperstack– direct path to scale up with existing

technologies in 2-D format

• Suspension bioreactor using microcarriers– 3-D footprint targeting large scale stirred tank

bioreactors mimicking MAb technology

• Disposable patient designated options– Quantum Hollow Fiber Bioreactor (Terumo)

– Xpansion Bioreactor (ATMI)

Heavy Lifting Scale UP

Courtesy of Jon Rowley, Lonza

Prohibitive Development Costs for Large Scale Manufacturing

• Individual companies cannot afford to perform large scale process development

• Catch-22 – bioprocessing industry is only now recognizing market opportunity for investment in cell therapy tools

• Industry consortia (ISCT, ARM) being organized for public domain development tools emphasizing downstream processing

Major Process Changes

15

Quantum

Cell Expansion System from CaridianBCT

• An automated integrated cell culture system

Bioreactor, incubator, media and waste management

• Controlled, reproducible cell expansion

• Decrease labor

• Self-contained bench top unit using a hollow fiber bioreactor

16

• MultiStem® was seeded in parallel on the Quantum and in standard culture conditions on cell culture plastic

• Growth is monitored and feed rates are adjusted based on glucose and lactate levels

• QC assays

• Growth rates and viability at harvest and after freeze/thawing

• Cytogenetics

• Flow cytometric profile

• Differentiation potential

• Immunomodulatory properties

QC

17

• Identical growth rates

• Viability at Quantum harvest: 93.9 % ± 1.6 %

• Viability after freeze/thawing: 90.1 % ± 2.2 %

Growth and viability

Quantum

18

Growth

With density centrifugation Whole BM load

Whole BM load on Quantum as good as seeding BMMNC on

cell culture plastic

19

Quantum

Isotype

Marker

• Panel of 22 markers was tested

Flow cytometric profile

Marker CriteriumStandardculture Quantum

CD3 < 1% 0.03% 0.14%

CD13 > 95% 97.51% 99.45%

CD19 < 1% 0.04% 0.03%

CD34 < 1% 0.00% 0.00%

CD40 < 1% 0.07% 0.03%

CD44 > 95% 98.51% 97.86%

CD45 < 1% 0.05% 0.03%

CD49c > 95% 94.47% 98.71%

CD49d N/A 11.34% 2.79%

CD54 N/A 6.18% 10.60%

CD56 < 1% 0.08% 0.08%

CD58 N/A 65.90% 72.68%

CD73 > 95% 97.85% 99.38%

CD80 < 1% 0.02% 0.01%

CD86 < 1% 0.38% 0.09%

CD90 > 95% 99.91% 99.92%

CD105 > 95% 99.53% 95.76%

CD117 N/A 28.49% 27.75%

CD140b < 1% 0.07% 0.10%

CD146 N/A 97.55% 92.15%

HLA I < 25% 5.39% 6.20%

HLA II < 1% 0.01% 0.01%

No significant differences for the flow cytometric markers and no pass/fail

criteria are exceeded

CD13HLA II CD49c

Standard conditions

Isotype

Marker

20

• Cytogenetic analysis on Agilent CGH arrays

Cytogenetics

No significant differences in karyotype

21

• Multistem® was differentiated to osteoblasts and adipocytes

• Histochemistry

Differentiation potential

22

Angiogenesis assay

L07 CES

L09 CES

23

• Quantification of differentiation

• qPCR

• Flow cytometry

Differentiation potential

CONFIDENTIAL 24

Maintenance of biological properties

Epigenetics and Profiling Comparability Testing

TranscriptomeProteome

miRNAGene methylation

MultiStem: A Distinctive ProductMultivariate analysis of Gene Expression Profile

Violet - MSCGreen - MesoangioblastsRed - MAPC (Verfaillie lab)Blue - MultiStem

26

MSC / likeMAPC / MultiStem

Violet - MSCGreen - MesoangioblastsRed - MAPC (Verfaillie lab)Blue - MultiStemOrange - ESC

ESC

MAPC / MultiStem

MSC / like

CONFIDENTIAL

• Key miRNA markers (MS vs MSC)

27

miRNAs as markers

MultiStem upregulated miRNAs(MS markers)

MultiStem downregulated miRNAs(MSC markers)

qPCR profiling Microarray profiling qPCR profiling Microarray profiling

MS markers

miR-1

miR-2

miR-3

miR-4

miR-5

miR-6

miR-7

miR-8

miR-9

miR-10

MSC markers

miR-a

miR-b

miR-c

miR-d

miR-e

miR-f

high stringency : FC > 1,5 in all three donors

low stringency : FC> 1,5 in at least two donors

Stability of Gene Methylation Profiles at Early and Late Expansion Stages

• Methylation profile of early (~20 PD) and late (~40 PD) MultiStem cultures depicted as the ratio of early/late methylation levels for 1536 CpGislands

• x-axis represents the Gene ID corresponding to the alphabetical listing of the genes associates with the CpG islands as shown in supplemental data

• Average ratio for two independent MultiStem lines, BMC2 and BMC 3 was 1.08+0.31 and 1.25+0.48, respectively

28

See Boozer, S, Lehman N, Lakshmipathy et al. Global Characterization and Genomic Stability of Human MultiStem, A Multipotent Adult Progenitor Cell. J Stem Cells 2009: 4(1).

Confidential

CONFIDENTIAL 29

Distribution of methylation cores

CONFIDENTIAL 30

Analysis overview

‘MultiStem miRNAs’ ‘MSC miRNAs’

Predicted miRNA targets

‘MultiStem mRNAs’ ‘MSC mRNAs’

OVERLAP ?

mRNAs that are predicted to be a

target, based on the presence of

conserved seed-sequences in the 3’-

UTR (via TargetScan)

Xeno-free Workflow

32

• Identical growth as in MS medium when cells are isolated in TheraPEAK™ on Synthemax coatings and subsequently transferred to CellBIND coatings

Surface Coatings are Critically Linked to Media Formulation

days

PD

0 10 20 30 40 500

5

10

15

20

25

MS

TP Synthemax

TP Synthemax -> CellBIND

TP CellBIND

MS IVTTheraPEAK

Morphology 9 days after seeding

CONFIDENTIAL 33

2D-DIGE workflow

Derivatise withCy3

Derivatise withCy5

Spots too dim to view by eye

Overlayimages

Mix and separate on one single

2D gel

Isoelectric focusingSDS-PAGE

E Cy3Image A

E Cy2Image C

DeCyder 2-D DifferentialAnalysis Software

Control Test

False-coloured pattern of differential spots is visible

Derivatise withCy2

Acquire images fromthe gel

E Cy5Image B

Cy3-labelledsample

Cy5-labelledsample

Cy2-labelledsample

Internal standard

CONFIDENTIAL 34

Gel 1: Control 1 Cy3 + Test 1 Cy5 + Internal standard Cy2Gel 2: Control 1 Cy5 + Test 1 Cy3 + Internal standard Cy2

Gel 3: Control 2 Cy3 + Test 2 Cy5 + Internal standard Cy2Gel 4: Control 2 Cy5 + Test 2 Cy3 + Internal standard Cy2

Gel 5: Control 3 Cy3 + Test 3 Cy5 + Internal standard Cy2Gel 6: Control 3 Cy5 + Test 3 Cy3 + Internal standard Cy2

CONFIDENTIAL 35

-PCA analysis: on pH 4-7, all spots, filtered only on: “spots have to be present in >75% of the spot maps”

PC

2

4.8

4.6

4.4

4.2

4

3.8

3.6

3.4

3.2

3

2.8

2.6

2.4

2.2

2

1.8

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

-0.2

-0.4

-0.6

-0.8

-1

-1.2

-1.4

-1.6

-1.8

-2

-2.2

-2.4

-2.6

-2.8

-3

-3.2

-3.4

-3.6

-3.8

-4

-4.2

-4.4

-4.6

-4.8

PC1876543210-1-2-3-4-5-6-7-8

MultiStem medium

TheraPEAK medium

Donor SVG/2

Donor L11

Secretome Following Immune Priming: Stem Cells Translational Medicine (Burrows et al)

Acknowledgments

Bart VaesDavid Craeye, Annelies Bogaerts, Silvie Cloosen, Saskia van Cruchten Saartje Walbers, Peter Sterkendries, Marian CrabbéKristel Gijbels, Ellen Van Houtven

www.regenesys.eu